WO2011148556A1 - 画像符号化装置、画像符号化方法、プログラムおよび集積回路 - Google Patents
画像符号化装置、画像符号化方法、プログラムおよび集積回路 Download PDFInfo
- Publication number
- WO2011148556A1 WO2011148556A1 PCT/JP2011/001926 JP2011001926W WO2011148556A1 WO 2011148556 A1 WO2011148556 A1 WO 2011148556A1 JP 2011001926 W JP2011001926 W JP 2011001926W WO 2011148556 A1 WO2011148556 A1 WO 2011148556A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- image
- unit
- parallax
- region
- images
- Prior art date
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/46—Embedding additional information in the video signal during the compression process
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N13/10—Processing, recording or transmission of stereoscopic or multi-view image signals
- H04N13/106—Processing image signals
- H04N13/128—Adjusting depth or disparity
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N13/10—Processing, recording or transmission of stereoscopic or multi-view image signals
- H04N13/106—Processing image signals
- H04N13/144—Processing image signals for flicker reduction
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N13/10—Processing, recording or transmission of stereoscopic or multi-view image signals
- H04N13/106—Processing image signals
- H04N13/161—Encoding, multiplexing or demultiplexing different image signal components
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
- H04N19/597—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding specially adapted for multi-view video sequence encoding
Definitions
- the present invention relates to an image encoding device that performs compression encoding of a stereoscopic video signal, and more particularly to an image recording device or an image such as a digital video camera or a digital still camera that handles two video signals with parallax, that is, a stereo video signal.
- the present invention relates to an image encoding device that is used in transmission equipment and encodes a stereo image.
- a conventional image encoding device that encodes a stereoscopic video signal (stereo video signal) controls the amount of code by controlling the degree of quantization according to the parallax of the stereo video signal. That is, in the conventional image coding device, when the parallax is small, the importance of the area having the parallax is large. Therefore, the information amount of the area is increased, and when the parallax is large, the area having the parallax is large. The amount of information is reduced, and as a result, processing for blurring the image in that region is performed (see, for example, Patent Document 1).
- FIG. 1 is a block diagram showing a configuration of a conventional image encoding device described in Patent Document 1.
- the video signals from the two cameras are input to the respective process circuits 101 and 102 to obtain RGB component signals.
- the RGB component signals are converted into digital data strings by the AD conversion units 103 and 104 and then stored in the memories 105 and 106.
- Each of the memories 105 and 106 is a memory that can store digital data strings for 8 lines. Eight lines of digital data strings, that is, data of an area (block) composed of pixels of 8 rows ⁇ 8 columns on the screen are read from the memories 105 and 106, respectively. Subsequent processing is performed for each block.
- the block data stored in the memory 105 is input to the DCT circuit 107, where it is subjected to discrete cosine transform and converted to a coefficient block composed of real number data in the frequency domain.
- This coefficient block is input to the quantization circuit 108, where it is multiplied by the inverse of a predetermined quantization step to be converted into an integer.
- the quantization step is determined by utilizing human visual characteristics, and is set so as to allow less distortion on the lower frequency side and allow more distortion on the higher frequency side. That is, the quantization step for the low frequency side coefficient is set small, and a large amount of code is allocated to the low frequency side.
- zero run-length encoding by the zero pack circuit 109 is performed on the data quantized in this way. That is, the zero pack circuit 109 encodes the quantized data by counting the number of consecutive zeros and combining the number with the coefficient that stops the consecutive zeros.
- the image encoding device described in Patent Document 1 includes a subtractor 110 that performs subtraction processing for each pixel of blocks stored in the memories 105 and 106 for each pixel, and a subtraction result obtained by the subtracter 110. And an absolute value sum circuit 111 for obtaining a sum of absolute values (parallax signal).
- the absolute value sum of the data difference for each pixel obtained in the absolute value sum circuit 111 corresponds to an image shift, that is, parallax, for the block.
- the quantization circuit 108 adjusts the quantization step according to the parallax signal that is the output of the absolute value sum circuit 111.
- the Huffman encoding unit 112 performs Huffman encoding, which is one of entropy encoding, on the run-length encoded data output from the zero pack circuit 109.
- the image encoding device described in Patent Document 1 increases the compression rate of a block with a large parallax, thereby increasing the compression rate of the block and consequently improving the encoding efficiency. At the same time, blurring is applied.
- FIG. 2A is a diagram showing a photographing method using two cameras.
- the cameras 21 and 22 photograph the object 23 in a state where the respective optical axes are crossed.
- the photographing method in such a state is called a crossing method.
- the optical axis is an axis in the center of an image obtained by photographing with a camera and along a direction perpendicular to the plane of the image.
- the parallax is a difference between the position of the image (left image) of the object 23 photographed by the camera 21 and the position of the image (right image) of the object 23 photographed by the camera 22.
- the distance from the cameras 21 and 22 along the direction perpendicular to the arrangement direction of the cameras 21 and 22 is referred to as a shooting distance. Further, the distances from the cameras 21 and 22 along the vertical direction to the intersection and the focal point are referred to as the intersection distance and the focal distance, respectively.
- the intersection distance and the focal distance are equal, the image of the object 23 is clear, and the parallax of the image is the smallest.
- the shooting distance is longer than the intersection distance (focal length)
- the parallax of the image of the distant object 24 at the shooting distance is indicated by the differences 25a and 25b and is sufficiently larger than the parallax of the object 23. .
- FIG. 2B is a diagram showing the relationship between the shooting distance and the parallax in the intersection method.
- the parallax D of the image of the subject at the shooting distance L decreases as a positive value as the shooting distance L increases if the shooting distance L is between 0 and the intersection distance L1. Further, the parallax D is 0 at the intersection distance. Further, when the shooting distance L is longer than the intersection distance L1, the parallax D decreases as a negative value as the shooting distance L increases. That is, in this case, the parallax D is a negative value, and the absolute value of the parallax D increases as the shooting distance L increases.
- the image of the object 23 has no parallax.
- the parallax of the image of the distant object 24 located farther from the cameras 21 and 22 than the intersection is large in the negative direction.
- the viewer can easily stereoscopically view the image of the object 23, but in order to stereoscopically view the image of the object 24 in a distant view, it is necessary to keep the eyes of both eyes apart from each other. There is. As a result, both eyes become tired.
- the image encoding apparatus of Patent Document 1 specifies the difference between the images of one block and the other block as the magnitude of the parallax, and increases the quantization step for a block with a large parallax. In this way, the image of the block having a large parallax is blurred. That is, the image encoding apparatus of Patent Document 1 reduces the amount of information in a region where the parallax is large because the importance of the region is low, and reduces the amount of information in the region where the parallax is small because the importance of the region is large. There are many.
- Patent Document 1 has a problem that blurring processing may be performed on an important region.
- Such an encoding technique by the image encoding apparatus is a technique that can be applied only to the imaging method based on the intersection method shown in FIG. 2A and cannot be applied to other imaging methods.
- FIG. 3A is a diagram showing another photographing method using two cameras.
- the cameras 21 and 22 photograph the object 23 in a state where the respective optical axes are parallel.
- the photographing method in such a state is called a parallel method.
- the cameras 21 and 22 are focused on the object 23.
- the image of the object 23 is clear.
- the parallax of the image of the object 23 is indicated by the differences 26a and 26b, and the parallax of the image of the distant object 24 at a position where the shooting distance L is longer than the object 23 is indicated by the differences 27a and 27b.
- FIG. 3B is a diagram showing the relationship between the shooting distance and the parallax in the parallel method.
- the parallax D of the image of the subject at the shooting distance L is always a positive value as shown in FIG. 3B, and decreases as the shooting distance L increases.
- the image of the object 23 has parallax.
- the parallax of the image of the distant object 24 located farther from the cameras 21 and 22 than the focal point is smaller than the parallax of the image of the object 23.
- the image encoding device described in Patent Document 1 encodes video signals from the cameras 21 and 22 obtained by the parallel method, a small amount of information is allocated to a large parallax area, and the parallax is small. Since a large amount of information is assigned to a region, blur processing may be performed on an image of an important block (region) that is focused. Thus, the distribution of parallax differs depending on the shooting method, and the smaller the parallax, the more important the area is not necessarily.
- An object of the present invention is to provide an image encoding device capable of performing the above.
- an image encoding device is an image encoding device that encodes two images for stereoscopic viewing, and a method for capturing the two images by a camera.
- a method specifying unit that specifies the parallax, a parallax detection unit that detects a parallax between the two images for each region constituting each of the two images, and a shooting specified by the method specifying unit for each of the regions
- a method, an adjustment unit determines an adjustment amount for blurring an image based on the parallax detected for the region by the parallax detection unit, and an adjustment amount that is determined for the region for each region
- an encoding unit that encodes the two images so that the image of the region is blurred.
- an adjustment amount is determined based not only on the parallax but also on the shooting method.
- the two images are encoded so that they are blurred. By blurring the image in this way, it is possible to suppress fatigue due to the stereoscopic vision of the viewer.
- the adjustment amount is determined based not only on the parallax but also on the shooting method, it is possible to prevent blurring processing from being performed on an important region due to the shooting method.
- the method specifying unit specifies whether the imaging method is a parallel method or a crossing method, and the adjustment unit is configured to specify that the imaging method is specified to be a crossing method by the method specifying unit.
- the adjustment amount may be determined for a region where the parallax detected by the parallax detection unit is larger than a first threshold.
- the shooting method is a crossing method
- the disparity in the distant view region is relatively large. Therefore, the adjustment amount is determined for the region where the disparity is greater than the first threshold value, so that the distant view image with a large disparity is determined. Can be appropriately blurred, and fatigue due to the stereoscopic vision of the viewer can be suppressed.
- the intersection method the parallax in the area where the subject at the intersection of the optical axes closer to the distant view, that is, the important area is extremely small, and the adjustment amount is not determined for the area where the parallax is small. It is possible to prevent the blurring process from being performed on a heavy region.
- the adjustment unit may determine an adjustment amount such that the greater the parallax, the greater the degree of blurring.
- the image encoding device further includes a focus specifying unit that specifies a focused region of each of the two images obtained by photographing with the camera, and the adjustment unit includes the focus You may set the magnitude
- the parallax in the focused area (so-called focus area) is set as the first threshold value, and the adjustment amount is determined for the area having the parallax larger than the first threshold value. This makes it possible to blur out distant images while improving the image quality.
- the method specifying unit specifies whether the imaging method is a parallel method or a crossing method, and the adjustment unit is configured to specify that the imaging method is specified by the method specifying unit as a parallel method. May determine the adjustment amount for a region where the parallax detected by the parallax detection unit is smaller than a second threshold.
- the shooting method is a parallel method
- the disparity in the region where the distant view is shown is relatively small. Can be properly blurred.
- the parallax in the area that shows the subject closer to the distant view, that is, the important area is relatively large, and the adjustment amount is not determined for the area where the parallax is large. It is possible to prevent the processing from being performed.
- the adjustment unit may determine an adjustment amount such that the degree of blurring increases as the parallax decreases.
- an image with a smaller parallax is blurred with stronger intensity, so that a distant view image can be blurred appropriately.
- the image encoding device further includes a focus specifying unit that specifies a focused region of each of the two images obtained by photographing with the camera, and the adjustment unit includes the focus You may set the magnitude
- the parallax in the focused area (so-called focus area) is set as the second threshold value, and the adjustment amount is determined for the area smaller than the second threshold value. In other words, it is possible to blur a distant view image while improving the image quality.
- the method specifying unit specifies an angle at which two optical axes of the camera for generating each of the two images intersect as a convergence angle, and whether or not the convergence angle is smaller than a predetermined angle.
- the imaging method is specified as a crossing method.
- the imaging method is specified as a parallel method. Good.
- the convergence angle is specified, and according to the convergence angle, whether the imaging method is the parallel method or the crossing method is specified, and thus the imaging method can be appropriately specified.
- the parallax detection unit detects a parallax between the two images as a parallax vector for each region
- the method specifying unit includes a distribution of parallax vectors detected for each region by the parallax detection unit. In this case, it is determined whether or not a negative-direction parallax vector and a positive-direction parallax vector are mixed, and if it is determined that they are mixed, the shooting method is identified as a crossing method and mixed. If it is determined that it is not, the photographing method may be specified as a parallel method.
- the imaging method is the parallel method or the intersection method, and therefore, for example, the convergence angle is set as compared with the case where the imaging method is specified based on the convergence angle. It is not necessary to specify bothersomely, and the configuration and processing operation of the image coding apparatus can be simplified.
- the encoding unit for each of the two images, a transform unit that performs orthogonal transform for each region, and a quantization that quantizes the frequency coefficient obtained by orthogonal transform by the transform unit for each region
- the quantization unit increases the quantization step by an adjustment amount determined by the adjustment unit for the region, and the increased quantum
- the image of the region may be blurred by quantizing the frequency coefficient of the region using the conversion step.
- the image is blurred due to an increase in the quantization step, so the parameters used for image coding can be diverted to the image blurring process, without adding processing or components for blurring the image.
- the configuration and processing operation of the image encoding device can be simplified.
- the encoding unit includes a transform unit that performs orthogonal transform for each of the two images, and the transform unit performs orthogonal transform on any region, and performs the transform on the region.
- the frequency coefficient included in the area is rounded down to a high frequency area having a width corresponding to the adjustment amount determined by the adjustment unit with respect to the area. You may blur the image.
- the image is blurred by truncating the frequency coefficient in the high frequency region, so that the image can be appropriately blurred using the data generated by encoding the image, and the processing and components for blurring the image can be performed.
- the configuration and processing operation of the image coding apparatus can be simplified without adding.
- the encoding unit generates a predicted image by predicting one of the two images from the other image using the parallax detected for each region, and the one image And calculating the difference between the predicted image and the predicted image.
- the method specifying unit specifies an angle between lines indicating the imaging directions of the two images as the imaging method, and the adjustment unit is specified by the method specifying unit for each region.
- the adjustment amount for blurring the image is determined based on the obtained angle and the parallax detected for the region by the parallax detection unit.
- the image is blurred according to the angle between the so-called center line of each of the two images and the parallax, so that it is possible to prevent blurring processing from being performed on an important region due to the angle. it can.
- an image encoding apparatus is an image encoding apparatus that encodes two images for stereoscopic viewing, and is obtained by parallel method imaging using a camera.
- a parallax detection unit that detects the parallax between the generated two images for each region constituting each of the two images, and the parallax detected for each region by the parallax detection unit with respect to the region
- an adjustment unit that determines an adjustment amount for blurring the image, and for each region, the two images are encoded so that the image in the region is blurred by the adjustment amount determined for the region.
- an encoding unit Accordingly, it is possible to prevent the blurring process from being performed on an important region when the photographing method is the parallel method.
- the present invention can be realized not only as such an image encoding apparatus, but also stores a method of processing operation in the image encoding apparatus, a program for causing the image encoding apparatus to perform processing operation, and the program.
- the present invention can also be realized as a recording medium, a system including the image encoding device, and an image decoding device that decodes a signal encoded by the image encoding device.
- the image encoding apparatus can suppress fatigue due to stereoscopic viewing of a viewer and can prevent blurring processing from being performed on an important region due to a photographing method.
- FIG. 1 is a block diagram showing a configuration of a conventional image encoding device.
- FIG. 2A is a diagram illustrating a photographing method using two cameras.
- FIG. 2B is a diagram illustrating the relationship between the shooting distance and the parallax in the intersection method.
- FIG. 3A is a diagram illustrating another imaging method using two cameras.
- FIG. 3B is a diagram illustrating the relationship between the shooting distance and the parallax in the parallel method.
- FIG. 4 is a block diagram showing a configuration of the image coding apparatus according to Embodiment 1 of the present invention.
- FIG. 5A is a diagram showing a range in which blurring processing is performed in the above intersection method.
- FIG. 5B is a diagram illustrating a range where blurring processing is performed in the parallel method.
- FIG. 6 is a flowchart showing the operation of the above-described image encoding apparatus.
- FIG. 7 is a flowchart showing the details of the blurring degree adjustment process.
- FIG. 8 is a block diagram showing a configuration of an image encoding device according to the modification.
- FIG. 9A is a diagram illustrating a range where blurring processing is performed in the intersection method according to the modification example.
- FIG. 9B is a diagram illustrating a range in which blurring processing is performed in the parallel method according to the modified example.
- FIG. 9C is a diagram illustrating another range in which the blurring process is performed in the parallel method according to the modified example.
- FIG. 9D is a diagram showing an image of a subject in focus in the parallel method according to the modification example.
- FIG. 9A is a diagram illustrating a range where blurring processing is performed in the intersection method according to the modification example.
- FIG. 9B is a diagram illustrating a range in which blurring processing is performed in the parallel method according
- FIG. 10 is a block diagram showing the configuration of the image coding apparatus according to Embodiment 2 of the present invention.
- FIG. 11 is a block diagram showing a configuration of an image encoding device according to the modification example.
- FIG. 12 is a block diagram showing a configuration of an image coding system according to Embodiment 3 of the present invention.
- FIG. 13 is a block diagram showing the configuration of the image decoding apparatus according to Embodiment 4 of the present invention.
- FIG. 14 is a block diagram showing the configuration of the image decoding apparatus according to Embodiment 5 of the present invention.
- FIG. 15 is a block diagram showing a configuration of an image encoding device according to the present invention.
- FIG. 16 is a flowchart showing the operation of the image coding apparatus according to the present invention.
- FIG. 17 is a block diagram showing another configuration of the image encoding device according to the present invention.
- FIG. 18A is a diagram for explaining a configuration of a camera according to the present invention.
- FIG. 18B is a diagram for explaining another configuration of the camera according to the present invention.
- FIG. 4 is a block diagram showing a configuration of the image coding apparatus according to Embodiment 1 of the present invention.
- Stereo imaging device 300A captures a subject so that it can be viewed stereoscopically, and encodes a stereo video signal generated by the imaging.
- Such a stereo imaging device 300A includes the image encoding device 300 according to the first embodiment, cameras Ca and Cb, and a storage unit Me.
- the cameras Ca and Cb are arranged at a distance of, for example, 6.5 cm from each other (an average distance between human eyes) and are perpendicular to the plane including the cameras Ca and Cb (for example, on the paper surface of FIG. 1). It is attached so as to be rotatable around an axis along a vertical direction).
- Each of the cameras Ca and Cb captures a subject and outputs a video signal obtained by capturing to the image encoding device 300.
- the video signals output from the cameras Ca and Cb are collectively referred to as a stereo video signal.
- An image indicated by the video signal output from the camera Ca is referred to as a main image
- an image indicated by the video signal output from the camera Cb is referred to as a sub-image.
- the subject image is stereoscopically viewed by the main image and the sub-image.
- the video signal is composed of a plurality of pictures, and each of the cameras Ca and Cb sequentially generates and outputs pictures at substantially the same timing (
- the storage unit Me is a recording medium for storing image data (a local decoded image signal described later) output from the image encoding device 300.
- the image encoding device 300 generates and outputs an encoded stereo video signal by encoding the stereo video signal output from the cameras Ca and Cb.
- the image encoding device 300 encodes the video signals for each block (area) constituting a picture.
- the image encoding device 300 encodes each picture included in the video signal as an I picture, a P picture, and a B picture, respectively.
- the image coding apparatus 300 performs inter-picture prediction coding (motion compensation prediction coding).
- the image encoding device 300 when encoding the video signal output from the camera Cb, the image encoding device 300 performs disparity compensation prediction encoding and encodes each picture included in the video signal as a P picture. That is, the image encoding apparatus 300 predicts a picture of a sub-image generated at the same timing as the picture from the picture of the main image, and encodes the picture of the sub-image based on the prediction result.
- Such an image coding apparatus 300 includes a convergence angle adjustment unit 302, a coding adjustment unit 303, a selector 304, a detection unit 305, a subtractor 306, a switch 307, a conversion unit 308, a quantization unit 309, and a variable length coding unit. 310, an inverse quantization unit 311, an inverse transform unit 312, an adder 313, a switch 314, and a compensation unit 316.
- the convergence angle adjustment unit 302 identifies the convergence angles ⁇ of the cameras Ca and Cb.
- cameras Ca and Cb are each equipped with a sensor that detects an angle. These sensors detect the orientation of the camera Ca or Cb that is variably set by the above-described rotation as the above-described angle.
- the convergence angle adjusting unit 302 acquires the angles detected by these sensors, and specifies the convergence angle ⁇ based on these angles.
- the convergence angle adjustment unit 302 determines whether the imaging method using the cameras Ca and Cb is the parallel method or the cross method based on the convergence angle ⁇ , and encodes and adjusts the imaging method signal indicating the determination result.
- the data is output to the unit 303. For example, if the convergence angle ⁇ is smaller than a predetermined angle, the convergence angle adjustment unit 302 determines that the imaging method is the parallel method, and if the convergence angle ⁇ is equal to or greater than the predetermined angle, the imaging method Is determined to be an intersection method.
- the selector 304 acquires video signals output from the cameras Ca and Cb, switches the video signals alternately, and outputs them to the switch 307, the subtractor 306, or the detection unit 305. For example, the selector 304 performs switching for each picture. In this case, when the selector 304 acquires pictures from the cameras Ca and Cb at the same timing, the selector 304 first outputs the picture acquired from the camera Ca, and then outputs the picture acquired from the camera Cb. .
- the subtractor 306 calculates, for each block, the difference between the encoding target image indicated by the video signal output from the selector 304 and the prediction image indicated by the prediction signal output from the compensation unit 316.
- the subtractor 306 outputs a prediction error signal indicating the difference to the switch 307.
- the subtracter 306 calculates the above difference and outputs a prediction error signal.
- the switch 307 connects the selector 304 to the conversion unit 308 and transfers a video signal indicating the encoding target image from the selector 304 to the conversion unit 308.
- the switch 307 connects the subtractor 306 to the conversion unit 308 to predict the prediction error of the P picture or B picture. The signal is transferred from the subtractor 306 to the conversion unit 308.
- the conversion unit 308 acquires a video signal or a prediction error signal as an image signal via the switch 307 for each block, and performs orthogonal transform (for example, discrete cosine transform) on the image signal. Thereby, the conversion unit 308 converts the image signal into a frequency coefficient (DCT coefficient), that is, converts the area of the image signal from the spatial domain to the frequency domain, and outputs the frequency coefficient to the quantization unit 309.
- DCT coefficient frequency coefficient
- the quantization unit 309 acquires a frequency coefficient from the conversion unit 308 and performs quantization on the frequency coefficient. That is, the quantization unit 309 generates a quantized value by dividing the frequency coefficient by the quantization step.
- the quantization unit 309 receives the adjustment signal from the encoding adjustment unit 303, the quantization unit 309 increases the quantization step by the adjustment amount indicated by the adjustment signal.
- the quantization unit 309 is a quantization determined according to an image encoding standard (specifically, H.264 / AVC). The encoding target block is quantized using steps.
- the quantization unit 309 When the quantization unit 309 receives the adjustment signal for the encoding target block, the quantization unit 309 adds or multiplies the adjustment amount to the quantization step determined according to the image encoding standard. , Update the quantization step so that the quantization step increases. The quantization unit 309 quantizes the encoding target block using the increased quantization step.
- variable length encoding unit 310 performs variable length encoding on the quantized value generated by the quantization unit 309 and the vector (motion vector or disparity vector) output from the detection unit 305, thereby encoding encoded stereo video. Generate and output a signal.
- This variable length coding is a reversible coding.
- the inverse quantization unit 311 generates an inverse quantization frequency coefficient by performing inverse quantization on the quantized value generated by the quantization unit 309. That is, the inverse quantization unit 311 generates an inverse quantization frequency coefficient by multiplying the quantization value by the quantization step used in the quantization unit 309. Note that the inverse quantization frequency coefficient generated at this time includes a quantization error unlike the frequency coefficient generated by the transform unit 308.
- the inverse transform unit 312 performs inverse orthogonal transform (for example, inverse discrete cosine transform) on the inverse quantization frequency coefficient generated by the inverse quantization unit 311. Thereby, the inverse transform unit 312 transforms the inverse quantization frequency coefficient into an image signal, that is, transforms the region of the inverse quantization frequency coefficient from the frequency domain to the space domain, and outputs the image signal to the adder 313. To do.
- inverse orthogonal transform for example, inverse discrete cosine transform
- the adder 313 stores the image signal output from the inverse conversion unit 312 in the storage unit Me as a locally decoded image signal.
- the adder 313 compensates for the image signal output from the inverse conversion unit 312 and the switch 314 when the encoding target image is a P-picture or B-picture image (main image or sub-image).
- the prediction signal output from the unit 316 is added, and the addition result is stored in the storage unit Me as a local decoded image signal.
- the detection unit 305 obtains an encoding target image from the selector 304, and when the encoding target image is a main image of a P picture or a B picture, a motion vector is obtained for each block with respect to the encoding target image. To detect. That is, the detection unit 305 refers to a locally decoded image signal indicating another main image (I picture or P picture) that has been encoded and decoded and stored in the storage unit Me as a reference image. Then, the detection unit 305 detects a motion of the encoding target image as a motion vector by searching the reference image for a block that is the same as or similar to the encoding target image (encoding target block). Then, the detection unit 305 outputs the motion vector to the compensation unit 316 and the variable length coding unit 310.
- the detection unit 305 detects a disparity vector for each block with respect to the encoding target image. That is, the detection unit 305 refers to a locally decoded image signal indicating a main image (I picture, P picture, or B picture) that has been encoded and decoded and stored in the storage unit Me as a reference image. Then, the detection unit 305 searches the reference image for a block that is the same or similar to the encoding target image (encoding target block), and thereby determines the positional relationship between the encoding target block and the same or similar block. Is detected as a parallax vector (parallax).
- the picture of the sub-image including the encoding target block and the reference image (reference picture) as the main image are pictures generated at the same timing by the cameras Ca and Cb, respectively.
- the detection unit 305 outputs the parallax vector to the encoding adjustment unit 303, the compensation unit 316, and the variable length encoding unit 310.
- the disparity vector indicates the positional relationship between the blocks as a positional shift in the left-right direction. Further, either one of the left and right is a positive direction, and the other is a negative direction.
- the compensation unit 316 When the compensation unit 316 acquires a motion vector from the detection unit 305, the compensation unit 316 generates a prediction signal by performing motion compensation using the motion vector and the reference image, and outputs the prediction signal. That is, the same or similar blocks included in the reference image are spatially shifted according to the motion vector. On the other hand, when acquiring the parallax vector from the detection unit 305, the compensation unit 316 generates a prediction signal by performing parallax compensation using the parallax vector and the reference image, and outputs the prediction signal. That is, the same or similar blocks included in the reference image are spatially shifted according to the disparity vector.
- the switch 314 opens between the compensation unit 316 and the adder 313 when the encoding target image is a main image of an I picture.
- the switch 314 connects the compensation unit 316 to the adder 313 and adds the prediction signal from the compensation unit 316. Deliver to the container 313.
- the encoding adjustment unit 303 acquires the imaging method signal from the convergence angle adjustment unit 302 and acquires the parallax vector from the detection unit 305. Then, when the imaging method signal indicates the crossing method, the encoding adjustment unit 303 determines the magnitude of the disparity vector for a block in which a disparity vector having a magnitude greater than the first threshold and a negative direction is detected. An adjustment amount corresponding to the adjustment amount is determined, and an adjustment signal indicating the adjustment amount is output to the quantization unit 309. For example, the encoding adjustment unit 303 outputs an adjustment signal indicating a larger adjustment amount as the size of the parallax vector is larger. In this case, when the quantization unit 309 acquires the adjustment signal, the quantization unit 309 increases the quantization step by the adjustment amount indicated by the adjustment signal.
- the encoding adjustment unit 303 determines the magnitude of the disparity vector for a block in which a disparity vector having a magnitude smaller than the second threshold and a positive direction is detected. And an adjustment signal indicating the adjustment amount is output to the quantization unit 309. For example, the encoding adjustment unit 303 outputs an adjustment signal indicating a larger adjustment amount as the size of the parallax vector is smaller. In this case, when the quantization unit 309 acquires the adjustment signal, the quantization unit 309 increases the quantization step by the adjustment amount indicated by the adjustment signal.
- the image of the block is blurred as compared with an image encoded according to the above-described image encoding standard.
- a process for blurring an image is hereinafter referred to as a blurring process, and the process for determining the adjustment amount is referred to as a blurring degree adjustment process.
- FIG. 5A is a diagram showing a range where blurring processing is performed in the intersection method.
- the magnitude of the parallax vector is 0 at the intersection of the optical axes.
- the first threshold is 0, the direction of the parallax vector of the image of the subject at the shooting distance L far from the intersection is negative, and the magnitude thereof is 0 which is the first threshold. Greater than.
- the blurring process is performed on the image of the subject at the shooting distance L far from the intersection.
- FIG. 5B is a diagram showing a range where blurring processing is performed in the parallel method.
- the magnitude of the parallax vector at the distance La becomes the second threshold value.
- the direction of the parallax vector of the subject image at the shooting distance L farther than the distance La is positive, and the magnitude thereof is smaller than the second threshold.
- the blurring process is performed on the image of the subject at the shooting distance L farther than the distance La.
- FIG. 6 is a flowchart showing the operation of the image coding apparatus 300 in the present embodiment.
- the image encoding apparatus 300 encodes the pictures for each block.
- the image encoding device 300 first determines whether or not the encoding target image that is the block is a main image (step S100). For example, a control unit (not shown) provided in the image encoding device 300 determines.
- the above-described control unit should further perform motion compensation predictive encoding on the encoding target image. Is determined (step S102). For example, when encoding the encoding target image as a P picture or B picture block, the image encoding apparatus 300 determines that motion compensation prediction encoding should be performed, and encodes the encoding target image as an I picture block. When converting, it is determined that motion compensation prediction encoding should not be performed.
- control unit determines that motion compensation predictive coding should be performed (Y in step S102)
- control unit controls the switch 307 to connect the subtractor 306 to the conversion unit 308 and the switch 314 to control the compensation unit.
- 316 is connected to the adder 313.
- the detection unit 305 detects a motion vector for the block of the encoding target image (step S104). Furthermore, the compensation unit 316 generates a prediction signal for the block of the encoding target image by performing motion compensation using the detected motion vector (step S106). The subtractor 306 generates a prediction error signal by subtracting the image indicated by the prediction signal from the encoding target image (step S108).
- the transform unit 308 generates a frequency coefficient by performing orthogonal transform on the prediction error signal generated in step S108 (step S110), and the quantization unit 309 quantizes the frequency coefficient by quantizing the frequency coefficient. A value is generated (step S112). Then, the variable length coding unit 310 performs variable length coding on the quantized value (step S114).
- step S102 when the control unit determines that the motion compensation prediction encoding should not be performed in step S102 (N in step S102), the control unit controls the switch 307 to connect the selector 304 to the conversion unit 308 and control the switch 314. Thus, the gap between the compensation unit 316 and the adder 313 is opened.
- the transform unit 308 generates a frequency coefficient by performing orthogonal transform on the block of the encoding target image output from the selector 304 (step S110).
- the quantization unit 309 quantizes the frequency coefficient to generate a quantized value (step S112), and the variable length coding unit 310 performs variable length coding on the quantized value (step S114).
- control unit determines that the image to be encoded is a sub-image in step S100 (N in step S100)
- the control unit controls the switch 307 to connect the subtractor 306 to the conversion unit 308 and switch the switch 314.
- the compensator 316 is connected to the adder 313 under control.
- the detection unit 305 detects a disparity vector for the block of the encoding target image (step S116). Further, the compensation unit 316 generates a prediction signal for the block of the encoding target image by performing parallax compensation using the detected parallax vector (step S118). The subtractor 306 generates a prediction error signal by subtracting the image indicated by the prediction signal from the encoding target image (step S120).
- the encoding adjustment unit 303 performs a blurring degree adjustment process based on the disparity vector detected in step S116 and the imaging method signal output from the convergence angle adjustment unit 302 (step S130).
- the transform unit 308 generates a frequency coefficient by performing orthogonal transform on the prediction error signal generated in step S120 (step S140). Further, the quantization unit 309 generates a quantized value by quantizing the frequency coefficient using the quantization step (step S142).
- the quantization unit 309 increases the quantization step according to the adjustment amount, and the increased quantization step is determined. Use to quantize.
- the quantization unit 309 performs quantization using the quantization step determined according to the image encoding standard. .
- the variable length coding unit 310 performs variable length coding on the quantized value (step S142).
- FIG. 7 is a flowchart showing details of the blurring degree adjustment process (step S130).
- the convergence angle adjustment unit 302 identifies the convergence angle ⁇ , and determines whether the imaging method using the cameras Ca and Cb is the cross method or the parallel method based on the convergence angle ⁇ (step S132). If it is determined that the shooting method is the intersection method (intersection method in step S132), the encoding adjustment unit 303 determines the direction of the disparity vector detected by the detection unit 305 with respect to the block of the encoding target image. Is negative and it is determined whether or not the magnitude of the parallax vector is larger than the first threshold (step S134).
- the encoding adjustment unit 303 determines that the disparity vector Based on the magnitude, the adjustment amount of the quantization step used by the quantization unit 309 is determined (step S138). For example, the encoding adjustment unit 303 determines a larger adjustment amount as the size of the parallax vector is larger.
- an adjustment amount for example, 0 or 1
- step S132 when the convergence angle adjustment unit 302 determines that the imaging method is the parallel method (the parallel method in step S132), the encoding adjustment unit 303 detects the block for the encoding target image. It is determined whether the direction of the parallax vector detected by 305 is positive and the magnitude of the parallax vector is smaller than the second threshold (step S136).
- the encoding adjustment unit 303 determines that the disparity vector Based on the magnitude, the adjustment amount of the quantization step used by the quantization unit 309 is determined (step S138). For example, the encoding adjustment unit 303 determines a larger adjustment amount as the size of the parallax vector is smaller.
- the encoding adjustment unit 303 determines the adjustment amount. do not do. At this time, the encoding adjustment unit 303 may determine an adjustment amount (for example, 0 or 1) indicating that an increase in the quantization step is unnecessary as described above.
- the blurring process is performed only on the sub-image, but the blurring process may be performed on the main image and the sub-image.
- the image encoding apparatus 300 encodes the sub-image for each block after the main image, and after the adjustment amount for each block is determined, the adjustment amount is used to determine the sub-image. Similarly, the main image is encoded again.
- the shooting method is the intersection method
- the direction of the parallax vector detected with respect to the block is negative
- the magnitude of the parallax vector is larger than the first threshold value.
- a large quantization step is applied to the block.
- the block is encoded as a blurred image.
- the magnitude of the disparity vector of the block including the image of the intersection point of the intersection method is 0 (or substantially 0). Therefore, when the first threshold value is 0, the block to which the above-described large quantization step is applied is a region where a distant view where the shooting distance L is longer than the intersection is shown.
- the shooting method when the shooting method is the parallel method, the direction of the parallax vector detected with respect to the block is positive and the magnitude of the parallax vector is smaller than the second threshold value.
- a large quantization step is applied to the block.
- the block is encoded as a blurred image.
- the magnitude of the disparity vector of the block including the image of the subject at the distance La in the picture indicated by the video signal is the second threshold. Therefore, the block to which the above-described large quantization step is applied is a region in which a distant view where the shooting distance L is longer than the distance La is shown.
- the present embodiment it is possible to suppress the fatigue due to the stereoscopic vision of the viewer by the blurring process and to prevent the blurring process from being performed on an important area due to the photographing method.
- the present embodiment by identifying the shooting method based on the convergence angle, even for a disparity distribution that differs depending on the shooting method, high image quality for an important region, and blurring processing for a distant (background) region, Can be performed appropriately.
- the first threshold value is 0, but other values may be used.
- image encoding apparatus 300 uses fixed first and second threshold values, but the first and second threshold values may be changed.
- the encoding adjustment unit 303 receives a signal from the outside of the image encoding apparatus 300 and changes the first or second threshold according to the signal. Thereby, the above-mentioned important area can be changed.
- the image of the encoding target block is blurred by increasing the quantization step.
- the image of the block may be blurred by discarding the frequency coefficient in the high frequency region of the block.
- the encoding adjustment unit 303 outputs the adjustment signal described above to the conversion unit 308.
- the adjustment amount indicated by this adjustment signal is handled as the size of the high frequency region that is cut off by the conversion unit 308.
- the conversion unit 308 when acquiring the adjustment signal from the encoding adjustment unit 303, the conversion unit 308 has a high frequency having a width corresponding to the adjustment amount indicated by the adjustment signal among the frequency coefficients of the block generated by the orthogonal transformation.
- the frequency coefficient included in the region is rounded down. That is, the conversion unit 308 changes the frequency coefficient in the high frequency region in a wide range to 0 if the adjustment amount is large, and changes the frequency coefficient in the high frequency region in a narrow range to 0 if the adjustment amount is small.
- the image encoding apparatus adjusts the focal point, and appropriately blurs the image of the region where the distant view where the shooting distance L is longer than the focal point is reflected without blurring the image of the subject at the focal point. There is a feature in the point.
- FIG. 8 is a block diagram showing a configuration of an image encoding device according to this modification.
- the image encoding apparatus 500 further includes a focus adjustment unit 402 in addition to the above-described image encoding apparatus 300, and performs encoding adjustment instead of the encoding adjustment unit 303 of the image encoding apparatus 300. Part 503.
- the focus adjustment unit 402 adjusts the focus of the cameras Ca and Cb. At this time, the focus adjustment unit 402 performs adjustment so that the focal points (focal lengths) of the cameras Ca and Cb are the same. In addition, the focus adjustment unit 402 outputs a focus signal indicating the position of the block including the image of the focused subject in the pictures output from the cameras Ca and Cb to the encoding adjustment unit 503. For example, the focus adjustment unit 402 identifies the block by autofocus (specifically, a contrast detection method or the like).
- the encoding adjustment unit 503 When the encoding adjustment unit 503 acquires a focus signal for each picture, the magnitude of the disparity vector detected by the detection unit 305 for the block at the position indicated by the focus signal is used as the first or second threshold value. Set. That is, when the imaging method signal acquired from the convergence angle adjustment unit 302 indicates the intersection method, the encoding adjustment unit 503 sets the above-described parallax vector magnitude as the first threshold value. On the other hand, when the imaging method signal acquired from the convergence angle adjustment unit 302 indicates the parallel method, the encoding adjustment unit 503 sets the above-described parallax vector magnitude as the second threshold value.
- FIG. 9A is a diagram showing a range where blurring processing is performed in the intersection method.
- the first threshold is set to the magnitude of the parallax vector at the focal point. Then, the blurring process is performed on a block having a negative direction and a disparity vector whose magnitude is larger than the first threshold.
- the focal distance L may be shorter than the intersection.
- a position where a disparity vector having a negative direction and a magnitude equal to the first threshold value is referred to as a focus corresponding position, and a disparity vector of an image of a subject at a shooting distance L farther than the focus corresponding position.
- the direction is negative and its magnitude is greater than the first threshold.
- the blurring process is performed on the image (distant view) of the subject at the shooting distance L farther than the focus corresponding position.
- the blurring process may be performed on the image (distant view) of the subject at the shooting distance L far from the focal point instead of the focus corresponding position.
- a fourth threshold value (not shown) that is larger than the magnitude of the parallax vector at the focal point by a predetermined magnitude or ratio may be set.
- the direction of the parallax vector of the image of the subject at a shooting distance L that is closer to the focus and closer to the position corresponding to the fourth threshold is positive, and the magnitude thereof is Greater than the threshold of 4.
- the encoding adjustment unit 303 also determines an adjustment amount according to the magnitude of the disparity vector, even for a block in which a disparity vector whose magnitude is greater than the fourth threshold and whose direction is positive is detected, and the adjustment The adjustment signal indicating the amount is output to the quantization unit 309.
- the blurring process is also performed on an image of a subject such as a foreground at a shooting distance L closer to the focal point. Thereby, the fatigue of the viewer due to the large parallax of the foreground can be suppressed.
- FIG. 9B is a diagram showing a range where blurring processing is performed in the parallel method.
- the second threshold value is set to the magnitude of the parallax vector at the focal point. Therefore, in the parallel method, the direction of the parallax vector of the image of the subject at the shooting distance L far from the focus is positive, and the magnitude thereof is smaller than the second threshold value. As a result, the blurring process is performed on the image of the subject at the shooting distance L far from the focus.
- the image of the region (focus area) including the image of the subject at the focus can be sharpened, and the image of the region where the distant view other than the focus area can be blurred. it can. That is, there is a high possibility that the focal point and the intersection are shifted during general camera shooting, and a more natural three-dimensional image can be formed in accordance with the shooting composition intended by the photographer.
- FIG. 9C is a diagram showing another range where blurring processing is performed in the parallel method.
- the second threshold value may be set to a value that is smaller than the magnitude of the parallax vector at the focal point and close to 0 as much as possible.
- the direction of the parallax vector of the image of the subject (for example, an image of a distant background, etc.) that is far from the focus and is farther away from the position Lb corresponding to the second threshold is It is positive and its magnitude is less than the second threshold.
- blurring processing is performed on an image of a subject such as a background at a shooting distance L farther than the focus.
- a third threshold value that is larger than the magnitude of the parallax vector at the focal point by a predetermined magnitude or ratio may be set.
- the direction of the parallax vector of the subject image at a shooting distance L that is closer to the focus and closer to the position Lc corresponding to the third threshold is positive, and the magnitude thereof is Greater than the third threshold.
- the encoding adjustment unit 303 determines an adjustment amount corresponding to the magnitude of the disparity vector even for a block in which a disparity vector having a magnitude larger than the third threshold and a positive direction is detected.
- the adjustment signal indicating the amount is output to the quantization unit 309.
- the blurring process is also performed on an image of a subject such as a foreground at a shooting distance L closer to the focal point. Thereby, the fatigue of the viewer due to the large parallax of the foreground can be suppressed.
- parallax (parallax vector) is generated with respect to the image of the subject at the focus. Due to such parallax, the position of the subject image shown in each of the main image and the sub-image may be shifted, making it difficult to see the subject image.
- FIG. 9D is a diagram showing an image of a subject in focus.
- the main image pL and the sub-image pR show subject images dL and dR, respectively. Also, the subject is focused. At this time, the positions of the images dL and dR of the subjects are shifted due to the parallax (parallax vector) at the focal point. For example, the position of the subject image dL is shifted to the right from the center of the main image pL, and the position of the subject image dR is shifted to the left from the center of the sub-image pR.
- the image decoding device image reproduction device
- the stereoscopic image pLR obtained by stereoscopically viewing the main image pL and the sub-image pR includes the subject image dL.
- dR are displayed shifted to the left and right respectively. Therefore, the image decoding device (reproducing device) may display the stereoscopic image pLR ′ by shifting the images dL and dR of the subject so that the magnitude of the parallax vector becomes zero. Thereby, the image dLR of the subject in focus can be clearly displayed.
- the image coding apparatus 500 includes only one focus adjustment unit 402, but may include one focus adjustment unit 402 for each of the cameras Ca and Cb.
- the convergence angle adjustment unit 302 may rotate the cameras Ca and Cb. That is, the convergence angle adjustment unit 302 receives a signal indicating an angle (setting angle) to be set from the outside of the image encoding device, and rotates the cameras Ca and Cb so that the convergence angle ⁇ becomes the setting angle. You may let them. In this case, the convergence angle adjustment unit 302 specifies the set angle as the convergence angle ⁇ .
- the image coding apparatus includes only one convergence angle adjustment unit 302. However, the image coding apparatus may include one convergence angle adjustment unit 302 for each of the cameras Ca and Cb. Good.
- the image coding apparatus is characterized in that a shooting method is specified based on the distribution (disparity distribution) of disparity vectors detected for each block constituting a picture.
- FIG. 10 is a block diagram showing a configuration of the image coding apparatus according to Embodiment 2 of the present invention.
- the image coding apparatus 600 includes a parallax distribution identification unit 601, a coding adjustment unit 602, a selector 304, a detection unit 305, a subtractor 306, a switch 307, a conversion unit 308, a quantization unit 309, a variable length code.
- the same constituent elements as those included in the image encoding apparatus 300 according to the first embodiment are used. The same reference numerals are assigned and detailed description is omitted.
- the image coding apparatus 600 includes a parallax distribution identification unit 601 and a coding adjustment unit 602 instead of the convergence angle adjustment unit 302 and the coding adjustment unit 303 as compared with the image coding apparatus 300 of the first embodiment. I have.
- the image encoding device 600 generates a parallax distribution and specifies whether the imaging method is a parallel method or a crossing method based on the parallax distribution.
- the image encoding device 600 causes the cameras Ca and Cb to perform trial shooting of the main image and the sub image in order to generate a parallax distribution.
- the image encoding apparatus 600 encodes (orthogonal transform and quantization) an I picture included in the video signal output from the camera Ca, decodes (inverse orthogonal transform and inverse quantization), and The local decoded image signal is stored in the storage unit Me.
- the detection unit 305 acquires a picture included in the video signal output from the camera Cb, and detects a disparity vector of the block for each block constituting the picture. That is, the detection unit 305 detects the disparity vector of each block included in the sub-picture picture by referring to the above-described local decoded image signal of the I picture stored in the storage unit Me as a reference image.
- the disparity distribution identifying unit 601 acquires the disparity vector detected by the detecting unit 305, and generates a disparity vector distribution of each block included in the sub-picture picture, that is, a disparity distribution. Then, the parallax distribution identifying unit 601 determines whether or not a positive-direction parallax vector and a negative-direction parallax vector are mixed in the parallax distribution, and outputs a shooting method signal indicating the determination result.
- the data is output to the encoding adjustment unit 602. Specifically, when the parallax distribution identifying unit 601 determines that the shooting methods are mixed, the parallax distribution identification unit 601 outputs a shooting method signal indicating that the shooting method is a crossing method.
- the imaging method signal indicating that the imaging method is the parallel method is output.
- the encoding adjustment unit 602 acquires a shooting method signal from the parallax distribution identification unit 601. Similar to the encoding adjustment unit 303 of the first embodiment, the encoding adjustment unit 602 generates a disparity vector whose magnitude is larger than the first threshold and whose direction is negative when the imaging method signal indicates a crossing method. An adjustment amount corresponding to the magnitude of the parallax vector is determined for the block having the same, and an adjustment signal indicating the adjustment amount is output to the quantization unit 309. On the other hand, when the imaging method signal indicates the parallel method, the encoding adjustment unit 602 responds to the size of the disparity vector with respect to a block having a disparity vector whose size is smaller than the second threshold and whose direction is positive. The adjustment amount is determined, and an adjustment signal indicating the adjustment amount is output to the quantization unit 309.
- the sensor for detecting the angles of the cameras Ca and Cb, the convergence angle adjustment unit 302, and the like as in the first embodiment are required.
- the configuration and processing of the image encoding device can be simplified.
- the block image is blurred by increasing the quantization step.
- the block image is reduced by truncating the frequency coefficient in the high frequency region of the block. It may be blurred.
- the encoding adjustment unit 602 outputs the adjustment signal described above to the conversion unit 308.
- the adjustment amount indicated by this adjustment signal is handled as the size of the high frequency region that is cut off by the conversion unit 308.
- the transform unit 308 when acquiring the adjustment signal from the encoding adjustment unit 602, the transform unit 308 has a high frequency with a width corresponding to the adjustment amount indicated by the adjustment signal, among the frequency coefficients of the block generated by the orthogonal transform.
- the frequency coefficient included in the region is rounded down. That is, the conversion unit 308 changes the frequency coefficient in the high frequency region in a wide range to 0 if the adjustment amount is large, and changes the frequency coefficient in the high frequency region in a narrow range to 0 if the adjustment amount is small.
- the image coding apparatus is in a position where the shooting distance L is longer than the focus without adjusting the focus and blurring the image of the subject at the focus, as in the second modification of the first embodiment.
- the feature is that the image of the area where the distant view is reflected is appropriately blurred.
- FIG. 11 is a block diagram showing a configuration of an image encoding device according to this modification.
- the image encoding apparatus 700 further includes a focus adjustment unit 402 in addition to the above-described image encoding apparatus 600, and performs encoding adjustment instead of the encoding adjustment unit 602 of the image encoding apparatus 600. Part 702.
- the focus adjustment unit 402 adjusts the focus of the cameras Ca and Cb. At this time, the focus adjustment unit 402 performs adjustment so that the focal points (focal lengths) of the cameras Ca and Cb are the same. In addition, the focus adjustment unit 402 outputs, to the encoding adjustment unit 702, a focus signal indicating the position of the block including the focused subject image in the pictures output from the cameras Ca and Cb.
- the encoding adjustment unit 702 When the encoding adjustment unit 702 obtains a focus signal for each picture, the magnitude of the disparity vector detected by the detection unit 305 for the block at the position indicated by the focus signal is used as the first or second threshold value.
- the encoding adjustment unit 702 sets the above-described parallax vector magnitude as the first threshold value.
- the imaging method signal acquired from the parallax distribution identification unit 601 indicates the parallel method
- the encoding adjustment unit 702 sets the above-described magnitude of the parallax vector as the second threshold value.
- the image encoding apparatus 700 includes only one focus adjustment unit 402, but may include one focus adjustment unit 402 for each of the cameras Ca and Cb.
- FIG. 12 is a block diagram showing the configuration of the image coding system in the present embodiment.
- the image encoding system 800 includes an image encoding unit 801, a storage unit Me, a camera Ca, a camera Cb, a display unit 804, and an external recording device 805.
- Storage unit Me, camera Ca, and camera Cb in the present embodiment are the same as storage unit Me, camera Ca, and camera Cb in the first and second embodiments and their modifications.
- the image encoding unit 801 in the present embodiment is the image encoding device 300, 500, 600, or 700 in the first and second embodiments and their modifications.
- the camera Ca includes a lens Ca1, an image sensor Ca2, a main control unit Ca3, and a lens control unit Ca4.
- the image pickup device Ca2 is composed of, for example, a CCD (Charge Coupled Device), acquires an optical signal via the lens Ca1, converts the optical signal into an electrical signal, and outputs the electrical signal to the main controller Ca3.
- the lens control unit Ca4 adjusts the focal point of the lens Ca1 and the like according to the control from the main control unit Ca3.
- the main control unit Ca3 is composed of, for example, an IC (Integrated Circuit), acquires the electric signal output from the image sensor Ca2, and outputs the electric signal to the image encoding unit 801 as a video signal. Further, the main control unit Ca3 performs shutter speed, gain adjustment, focus adjustment, and the like by controlling the image sensor Ca2 and the lens control unit Ca4.
- the camera Cb includes a lens Cb1, an image sensor Cb2, a main control unit Cb3, and a lens control unit Cb4. These components included in the camera Cb are the same as the above-described components included in the camera Ca.
- the main control units Ca3 and Cb3 perform a cooperative operation so that the focal points and shutter speeds of the cameras Ca and Cb are the same.
- the display unit 804 includes a liquid crystal display, for example, acquires a stereo video signal from the image encoding unit 801, and displays a main image and a sub-image indicated by the stereo video signal. Note that the display unit 804 may acquire the local decoded image signal from the image encoding unit 801 and display the main image and the sub image indicated by the local decoded image signal.
- the external recording device 805 is configured to be loaded with a recording medium such as a CD (Compact Disc), an MO (Magnet Optical Disc), a DVD (Digital Versatile Disc), a BD (Blu-ray Disc), or a semiconductor memory. ing. Then, the external recording device 805 acquires the encoded stereo video signal from the image encoding unit 801 and writes the encoded stereo video signal to the recording medium attached.
- a recording medium such as a CD (Compact Disc), an MO (Magnet Optical Disc), a DVD (Digital Versatile Disc), a BD (Blu-ray Disc), or a semiconductor memory.
- FIG. 13 is a block diagram showing the configuration of the image decoding apparatus according to Embodiment 4 of the present invention.
- the image decoding apparatus 1000 is an apparatus that decodes the encoded stereo video signal output from the image encoding apparatus according to the first or second embodiment, or a modification thereof, and includes a variable length decoding unit 1001, an inverse quantization / inverse A conversion unit 1002, a motion / disparity compensation unit 1003, an adder 1004, a deblocking filter 1005, a filter processing identification unit 1006, an edge enhancement unit 1007, and a switch 1008 are provided.
- the image decoding apparatus 1000 decodes the main image and the sub-picture included in the encoded stereo video signal for each block while storing and reading data in the storage unit Me1, and enables stereoscopic viewing.
- the decoded picture is displayed on the display unit D1.
- the image decoding apparatus 1000 displays only the main image included in the encoded stereo video signal on the display unit D1. That is, when the display unit D1 does not support 3D video display, the image decoding apparatus 1000 displays the 2D video on the display unit D1.
- the variable length decoding unit 1001 performs variable length decoding on the encoded stereo video signal.
- the imaging method signal, the quantization value, the quantization step, and the motion vector or the disparity vector that have been variable-length encoded are decoded.
- the encoded stereo video signal output from the image encoding device according to the first or second embodiment or the modification thereof includes a shooting method signal that is variable-length encoded and a quantum that is variable-length encoded for each block. And a quantization step (quantization parameter) and a motion vector or a disparity vector that is variable-length-coded for each block.
- variable length decoding unit 1001 outputs a motion vector or a disparity vector to the motion / disparity compensation unit 1003, outputs a quantized value and a quantization step (quantization parameter) to the inverse quantization / inverse transform unit 1002, and captures a shooting method.
- the signal is output to the filter processing identification unit 1006.
- the inverse quantization / inverse transform unit 1002 uses the quantization step for the quantization value. Quantization is performed, and as a result, an inverse quantization frequency coefficient is generated for each block. Further, the inverse quantization / inverse transform unit 1002 performs inverse orthogonal transform (for example, inverse discrete cosine transform) on the inverse quantization frequency coefficient.
- inverse orthogonal transform for example, inverse discrete cosine transform
- the inverse quantization / inverse transform unit 1002 converts the inverse quantization frequency coefficient into an image signal, that is, converts the region of the inverse quantization frequency coefficient from the frequency domain to the spatial domain, and adds the image signal. Output to the device 1004. Further, the inverse quantization / inverse transform unit 1002 outputs the quantization step used for the inverse quantization for each block to the filter processing identification unit 1006.
- the adder 1004 When the decoding target image is a main image of an I picture, the adder 1004 outputs the image signal output from the inverse quantization / inverse transform unit 1002 to the deblocking filter 1005 as a decoded image signal. Further, when the decoding target image is a P-picture or B-picture image (main image or sub-image), the adder 1004 and the image signal output from the inverse quantization / inverse transform unit 1002 and the motion / disparity The prediction signal output from the compensation unit 1003 is added, and the addition result is output to the deblocking filter 1005 as a decoded image signal.
- the deblocking filter 1005 When the deblocking filter 1005 acquires the decoded image signal from the adder 1004, the deblocking filter 1005 filters the decoded image signal. That is, the deblocking filter 1005 removes distortion (blocking distortion) of the image of the block indicated by the decoded image signal, and stores the decoded image signal from which the distortion is removed in the storage unit Me1.
- the deblocking filter 1005 removes distortion (blocking distortion) of the image of the block indicated by the decoded image signal, and stores the decoded image signal from which the distortion is removed in the storage unit Me1.
- the motion / disparity compensation unit 1003 performs motion compensation using the motion vector acquired from the variable length decoding unit 1001 when the decoding target image (decoding target block) is a main image of a P picture or a B picture. A prediction signal is generated, and the prediction signal is output to the adder 1004. That is, the motion / disparity compensation unit 1003 refers to the decoded image signal indicating the already decoded main image (I picture or P picture) stored in the storage unit Me1 as a reference image. Then, the motion / disparity compensation unit 1003 spatially shifts the block in the reference image according to the motion vector, and generates and outputs a signal indicating the image of the block as a prediction signal.
- the motion / disparity compensation unit 1003 performs prediction by performing disparity compensation using the disparity vector acquired from the variable length decoding unit 1001. A signal is generated and the predicted signal is output to the adder 1004. That is, the motion / disparity compensation unit 1003 refers to the decoded image signal indicating the already decoded main image (I picture, P picture, or B picture) stored in the storage unit Me1 as a reference image. Then, the motion / disparity compensation unit 1003 spatially shifts the block in the reference image according to the parallax vector, and generates and outputs a signal indicating the image of the block as a prediction signal. Furthermore, the motion / disparity compensation unit 1003 outputs the disparity vector used for the disparity compensation to the filter processing identification unit 1006.
- the edge enhancement unit 1007 obtains the adjustment signal from the filter processing identification unit 1006, based on the adjustment signal, the edge enhancement unit 1007 performs a process of enhancing the edge of the image of the decoded image signal stored in the storage unit Me1, and the processing is performed.
- the decoded image signal is output to the switch 1008.
- Switch 1008 switches connection according to control by filter processing identification unit 1006. That is, the switch 1008 connects the display unit D1 to the storage unit Me1 via the edge enhancement unit 1007, or connects the display unit D1 to the storage unit Me1 without using the edge enhancement unit 1007.
- the filter processing identification unit 1006 controls the switch 1008,
- the display unit D1 is connected to the storage unit Me1.
- the decoded image signal indicating the main image and the sub image stored in the storage unit Me1 is read, and the main image and the sub image are alternately displayed on the display unit D1. Therefore, in this case, an image is displayed in a three-dimensional manner.
- the filter processing identifying unit 1006 when the filter processing identifying unit 1006 has acquired a main image display signal indicating that only the main image should be displayed, the filter processing identifying unit 1006 causes the switch 1008 to switch the connection for each block of the main image to be displayed. That is, for each block, the filter processing identification unit 1006 indicates whether or not the blurring process has been performed on the block at the time of encoding, and the disparity vector corresponding to the block output from the motion / disparity compensation unit 1003. The determination is made based on the imaging method signal output from the variable length decoding unit 1001. Then, the filter processing identification unit 1006 controls the switch 1008 according to the determination result.
- the filter processing identification unit 1006 when the imaging method signal indicates the intersection method, the filter processing identification unit 1006 has a negative parallax vector corresponding to the block and has a magnitude greater than the first threshold. Then, it is determined that the blur process has been performed on the block at the time of encoding. In addition, when the imaging method signal indicates the parallel method, the filter processing identification unit 1006 determines that the block when the direction of the disparity vector corresponding to the block is positive and the size is smaller than the second threshold value. It is determined that a blurring process has been performed during encoding.
- the filtering process identifying unit 1006 determines that the blurring process is not performed, the filtering process identifying unit 1006 controls the switch 1008 to connect the display unit D1 to the storage unit Me1. Thereby, the decoded image signal indicating the block of the main image stored in the storage unit Me1 is read, and the block of the main image is displayed on the display unit D1. On the other hand, when determining that the blurring process is performed, the filter processing identification unit 1006 controls the switch 1008 to connect the display unit D1 to the edge enhancement unit 1007.
- the edge enhancement unit 1007 reads out the decoded image signal indicating the block of the main image on which the blurring process has been performed from the storage unit Me1, emphasizes the edge of the main image (block) indicated by the decoded image signal, The block of the main image is displayed on the display unit D1.
- the filtering process identifying unit 1006 determines that the blurring process is performed, the filtering process identifying unit 1006 is based on the quantization step used for the inverse quantization of the block of the main image output from the inverse quantization / inverse transform unit 1002.
- the filter processing identification unit 1006 performs a predetermined quantization step (for example, a quantization step determined by the image coding standard) from the quantization step used for the inverse quantization of the main image block. Subtraction is performed, and the subtraction result is specified as the adjustment amount. Then, the filter processing identification unit 1006 outputs an adjustment signal indicating the adjustment amount to the edge enhancement unit 1007.
- the edge enhancement unit 1007 emphasizes the edge of the main image (block) indicated by the decoded image signal
- the edge enhancement unit 1007 acquires the adjustment signal from the edge enhancement unit 1007 and emphasizes the edge by the adjustment amount indicated by the adjustment signal. To do.
- the image decoding apparatus 1000 appropriately decodes the encoded stereo video signal generated by the image encoding apparatus according to the first or second embodiment or the modification thereof so as to be stereoscopically viewed. And can be displayed on the display unit D1. Furthermore, the image decoding apparatus 1000 according to the present embodiment can display only the main image on the display unit D1 when the display unit D1 does not support 3D video display. At this time, the image decoding apparatus 1000 appropriately emphasizes the edge of the blurred main image to make the main image clear 2D video even if the main image is blurred at the time of encoding. As the display portion D1.
- FIG. 14 is a block diagram showing the configuration of the image decoding apparatus according to Embodiment 5 of the present invention.
- the image decoding apparatus 1100 is an apparatus that decodes the encoded stereo video signal output from the image encoding apparatus according to the first or second embodiment, or a modification thereof, and includes a variable length decoding unit 1001, an inverse quantization / inverse A conversion unit 1002, a motion / disparity compensation unit 1003, an adder 1004, a deblocking filter 1010, a filter processing identification unit 1011, and a switch 1009 are provided.
- the image decoding apparatus 1100 decodes the main image and the sub-picture included in the encoded stereo video signal for each block while storing and reading data in the storage unit Me1, and can be stereoscopically viewed. Thus, the decoded picture is displayed on the display unit D1. Further, when the main image display signal is acquired from the outside, the image decoding device 1100 displays only the main image included in the encoded stereo video signal on the display unit D1. That is, when the display unit D1 does not support 3D video display, the image decoding apparatus 1000 displays the 2D video on the display unit D1.
- the encoded stereo video signal decoded in the present embodiment not only the sub-picture but also the main picture is subjected to the blurring process in the first and second embodiments or their modifications.
- the same constituent elements as those included in the image decoding apparatus 1000 according to the fourth embodiment are the same as the constituent elements according to the fourth embodiment.
- the same reference numerals are assigned and detailed description is omitted.
- Switch 1009 switches the connection according to control by the filter processing identification unit 1011. That is, the switch 1009 connects the adder 1004 to the input of the deblocking filter 1010, or connects the storage unit Me1 to the input of the deblocking filter 1010.
- the deblocking filter 1010 filters the decoded image signal, as in the deblocking filter 1005 of the fourth embodiment. That is, the deblocking filter 1010 removes the distortion (blocking distortion) of the image of the block indicated by the decoded image signal, and stores the decoded image signal from which the distortion is removed in the storage unit Me1.
- the deblocking filter 1010 acquires the adjustment signal from the filter processing identification unit 1011, the deblocking filter 1010 performs intensity filtering according to the adjustment amount indicated by the adjustment signal.
- the filter processing identification unit 1011 controls the switch 1009, Adder 1004 is connected to the input of deblocking filter 1010.
- the decoded image signal indicating the main image and the sub image stored in the storage unit Me1 is read, and the main image and the sub image are alternately displayed on the display unit D1. Therefore, in this case, an image is displayed in a three-dimensional manner.
- the filter processing identifying unit 1011 when the filter processing identifying unit 1011 acquires a main image display signal indicating that only the main image should be displayed, the filter processing identifying unit 1011 causes the switch 1009 to switch the connection for each block of the main image to be displayed. That is, the filtering process identifying unit 1011 determines whether or not the blurring process has been performed on each block at the time of encoding, similar to the filtering process identifying unit 1006 of the fourth embodiment. The determination is made based on the disparity vector corresponding to the block output from 1003 and the imaging method signal output from the variable length decoding unit 1001. Then, the filter processing identification unit 1011 controls the switch 1009 according to the determination result.
- the filtering process identifying unit 1011 determines that the blurring process is not performed, the filtering process identifying unit 1011 controls the switch 1009 to connect the adder 1004 to the input of the deblocking filter 1010.
- the decoded image signal indicating the main image block output from the adder 1004 is filtered by the deblocking filter 1010, output to the display unit D1 via the storage unit Me1, and displayed on the display unit D1. .
- the filter processing identifying unit 1011 determines that the blurring process is being performed, the filter processing identifying unit 1011 first controls the switch 1009 to connect the adder 1004 to the input of the deblocking filter 1010.
- the decoded image signal indicating the main image block output from the adder 1004 is filtered by the deblocking filter 1010 and stored in the storage unit Me1.
- the filter processing identification unit 1011 controls the switch 1009 to connect the storage unit Me1 to the input of the deblocking filter 1010.
- the above-described decoded image signal that has been filtered and stored in the storage unit Me1 is further filtered again by the deblocking filter 1010.
- the decoded image signal filtered again is output to the display unit D1 via the storage unit Me1, and displayed on the display unit D1.
- the filtering process identifying unit 1011 determines that the blurring process is being performed, the filtering process identifying unit 1011 is based on the quantization step used for the inverse quantization of the block of the main image output from the inverse quantization / inverse transform unit 1002.
- the filter processing identification unit 1011 performs a predetermined quantization step (for example, a quantization step determined by the image coding standard) from the quantization step used for inverse quantization of the block of the main image. Subtraction is performed, and the subtraction result is specified as the adjustment amount. Then, the filter processing identification unit 1011 outputs an adjustment signal indicating the adjustment amount to the deblocking filter 1010.
- the deblocking filter 1010 acquires the adjustment signal from the filter processing identification unit 1011 and performs filtering of the intensity according to the adjustment amount indicated by the adjustment signal.
- the image decoding apparatus 1100 appropriately decodes the encoded stereo video signal generated by the image encoding apparatus according to the first or second embodiment or the modification thereof so as to be stereoscopically viewed. And can be displayed on the display unit D1. Furthermore, the image decoding apparatus 1100 according to the present embodiment can display only the main image on the display unit D1 when the display unit D1 does not support 3D video display. At this time, the image decoding apparatus 1100 appropriately removes the distortion of the blurred main image even if the main image is subjected to the blurring process at the time of encoding, so that the main image becomes a clear 2D video. As the display portion D1.
- the image coding apparatus, the image coding system, and the image decoding apparatus according to the present invention have been described using Embodiments 1 to 5 and their modifications, but the present invention is not limited to these. It is not a thing.
- the image coding apparatus according to the present invention does not have to include all of the components shown in FIGS. 4, 8, 10 and 11, and may not include any of the components (for example, The components of FIG. 15 or FIG. 17 alone can achieve the effects of the present invention.
- FIG. 15 is a block diagram showing a configuration of an image encoding device according to the present invention.
- An image encoding device 10 is an image encoding device that encodes two images for stereoscopic viewing, and a method specifying unit that specifies a method for capturing two images by two cameras Ca and Cb. 12 and a parallax detection unit 11 that detects the parallax between the two images for each region that constitutes each of the two images, a shooting method that is identified by the method identification unit 12 for each region, and parallax detection
- An adjustment unit 13 that determines an adjustment amount for blurring an image based on the parallax detected for the region by the unit 11, and an adjustment amount determined for the region for each region.
- An encoding unit 14 that encodes two images is provided so that the images are blurred.
- the parallax detection unit 11 corresponds to the detection unit 305 in the first to fifth embodiments and their modifications.
- the method specifying unit 12 corresponds to the convergence angle adjusting unit 302 or the parallax distribution identifying unit 601 in the first to fifth embodiments and the modifications thereof.
- the adjustment unit 13 corresponds to the coding adjustment units 303, 503, 602, and 702 in the first to fifth embodiments and their modifications.
- Encoding section 14 corresponds to conversion section 308, quantization section 309, and variable length encoding section 310 in the first to fifth embodiments and their modifications.
- the image coding method according to the present invention does not need to include all of the steps shown in FIGS. 6 and 7, and does not include any of the steps (for example, only the steps shown in FIG. 16). However, the effects of the present invention can be achieved.
- FIG. 16 is a flowchart showing an image encoding method according to the present invention.
- the image encoding method according to the present invention is an image encoding method for encoding two images for stereoscopic viewing, and specifies a method for capturing two images by two cameras (step S10).
- the parallax between the two images is detected for each area constituting each of the two images (step S12), and for each of the areas, based on the imaging method specified and the parallax detected for the area,
- An adjustment amount for blurring the image is determined (step S14), and for each region, the two images are encoded so that the image in the region is blurred by the adjustment amount determined for the region (step S16).
- Encoding by such an image encoding method is performed by the image encoding device 10 shown in FIG.
- the image encoding device 10 and the image encoding method thereof according to the present invention when two images for stereoscopic viewing (for example, a main image and a sub image) are encoded, only parallax is used. Instead, the adjustment amount is determined based on the shooting method, and the two images are encoded so that the image is blurred by the adjustment amount. By blurring the image in this way, it is possible to suppress fatigue due to the stereoscopic vision of the viewer. Furthermore, since the adjustment amount is determined based not only on the parallax but also on the shooting method, it is possible to prevent blurring processing from being performed on an important region due to the shooting method.
- the encoding in the encoding unit 14 is not limited to the encoding (orthogonal transformation, quantization, variable length encoding, etc.) in the first to fifth embodiments and their modifications. .
- FIG. 17 is a block diagram showing another configuration of the image encoding device according to the present invention.
- An image encoding device 10a is an image encoding device that encodes two images for stereoscopic viewing, and a method specifying unit that specifies a method for capturing two images by two cameras Ca and Cb. 12 and a parallax detection unit 11 that detects the parallax between the two images for each region that constitutes each of the two images, a shooting method that is identified by the method identification unit 12 for each region, and parallax detection Based on the parallax detected for the region by the unit 11, an adjustment unit 13a that determines an adjustment amount for blurring the image, and for each region, the adjustment amount determined for the region An encoding unit 14 that encodes two images is provided so that the images are blurred.
- the image encoding device 10a further includes a focus specifying unit 15 that specifies a focused area of each of two images obtained by photographing with two cameras, and the adjustment unit 13a has a focus.
- the magnitude of the parallax detected by the parallax detection unit 11 for the area specified by the specifying unit 15 is set as the first or second threshold.
- the focus identifying unit 15 corresponds to the focus adjusting unit 402 in the first to fifth embodiments and their modifications.
- the effects of the above-described image encoding device 10 can be achieved, and since the parallax in the focus area is set as the first or second threshold value, the focus area is clearly defined. That is, it is possible to blur a distant view image while improving the image quality. Therefore, among the constituent elements in the first to fifth embodiments and the modifications thereof, those other than the constituent elements included in the image encoding device 10a are not essential constituent elements.
- the image of the area in which the distant view (background) is shown is blurred, but conversely, a large amount of information is allocated to the focus area, and the focus area is improved in image quality. May be.
- the image of the region in which the distant view (background) is shown is blurred. For example, the image of the region in which the close-up view whose shooting distance L is closer to the focus is shown. You may blur, and you may blur the image of the area
- the adjustment amount according to the magnitude of the parallax vector is determined, but a fixed adjustment amount may be determined regardless of the magnitude of the parallax vector.
- the parallax compensation prediction coding is performed on the video signal (sub-picture) output from the camera Cb, but the parallax compensation prediction coding and the motion compensation prediction code are performed. You may perform. That is, the image encoding apparatus may use a compression encoding method that uses both parallax compensation prediction and motion compensation prediction as a multi-view profile (MVC: Multi-View Profile).
- MVC Multi-View Profile
- stereo photographing apparatus 300A includes two cameras (Ca and Cb), but may include only one camera.
- the camera may have all the functions and configurations of the cameras Ca and Cb, or may include only one image sensor (for example, a CCD) for two optical devices (lenses).
- image sensor for example, a CCD
- the one image sensor receives light output from each of the two optical devices, and outputs an electrical signal indicating a main image and a sub-image. Convert to the electrical signal shown.
- the direction of the parallax vector does not change from positive to negative as the shooting distance L increases.
- the cameras (Ca and Cb) may be configured such that the direction of the parallax vector is changed as in the intersection method (as shown in FIG. 5A).
- 18A and 18B are diagrams for explaining the configuration of the camera.
- the optical axes of the cameras Ca and Cb intersect each other.
- an optical device and an image sensor corresponding to the optical device are arranged along the optical axis.
- the center line from the optical device of the camera Ca toward the center of the imaging region corresponding to the main image coincides with the optical axis of the optical device of the camera Ca.
- the center line from the optical device of the camera Cb toward the center of the shooting area corresponding to the sub-image coincides with the optical axis of the optical device of the camera Cb.
- these center lines are lines indicating the direction in which the main image or the sub-image is captured. Therefore, when the two optical axes intersect, the two center lines also intersect.
- the direction of the parallax vector changes from positive to negative as the shooting distance L increases as shown in FIG. 5A.
- the cameras Ca and Cb are arranged so that the optical axes of the cameras Ca and Cb are parallel to each other, photographing by the parallel method described above is performed. In this case, the center lines of the cameras Ca and Cb are also parallel. Therefore, as shown in FIG. 5B, the direction of the parallax vector does not change from positive to negative as the shooting distance L increases.
- the direction of the parallax vector changes from positive to negative as the shooting distance L increases as in the crossing method, that is, as shown in FIG. 5A.
- the imaging element is arranged so as to be shifted from the optical axis. That is, the cameras Ca and Cb are configured as so-called lens shift cameras.
- the center lines of the cameras Ca and Cb cross each other due to the displacement of the imaging device. As the center lines intersect, the direction of the parallax vector changes from positive to negative as the shooting distance L increases as shown in FIG. 5A.
- the image coding apparatus specifies the angle between the center lines as a shooting method, and the coding adjustment unit performs the operation for each block (region).
- An adjustment amount for blurring the image may be determined based on the specified angle and the parallax detected for the block.
- the center line is intersected in the parallel method by the lens shift camera.
- the center line may be intersected in the parallel method by another configuration.
- an optical device and an image sensor corresponding to the optical device are arranged along the optical axis, and the viewing angle of the optical device and the light reception of the image sensor.
- the area is set widely. Then, a part of the image generated by the image sensor (a part on the image side generated by the other image sensor) is trimmed as a main image or a sub-image. Even with such two cameras, the center lines can be crossed while being parallel, like the lens shift camera described above.
- the method for capturing two pictures (main image and sub-image) by the camera is specified, but the method for capturing the images may not be specified.
- the shooting method is a parallel method
- the image of the block is blurred by an adjustment amount corresponding to the parallax between two pictures for each block (area) without specifying the shooting method.
- Two pictures are encoded. For example, if the parallax is smaller than the threshold value, two pictures are encoded so that the image of the block corresponding to the parallax is blurred, and the image of the block corresponding to the parallax is strongly blurred as the parallax is small. Accordingly, it is possible to prevent the blurring process from being performed on an important region when the photographing method is the parallel method.
- the present invention includes the following cases.
- Each of the above devices is specifically a computer system including a microprocessor, a ROM, a RAM, a hard disk unit, a display unit, a keyboard, a mouse, and the like.
- a computer program is stored in the RAM or hard disk unit.
- Each device achieves its function by the microprocessor operating according to the computer program.
- the computer program is configured by combining a plurality of instruction codes indicating instructions for the computer in order to achieve a predetermined function.
- a part or all of the components constituting each of the above devices may be configured by one system LSI (Large Scale Integration).
- the system LSI is a super multifunctional LSI manufactured by integrating a plurality of components on one chip, and specifically, a computer system including a microprocessor, a ROM, a RAM, and the like. .
- a computer program is stored in the RAM.
- the system LSI achieves its functions by the microprocessor operating according to the computer program.
- each part of the constituent elements constituting each of the above devices may be individually made into one chip, or may be made into one chip so as to include a part or all of them.
- the system LSI is called here, but it may be called IC, LSI, super LSI, or ultra LSI depending on the degree of integration.
- the method of circuit integration is not limited to LSI's, and implementation using dedicated circuitry or general purpose processors is also possible.
- an FPGA Field Programmable Gate Array
- a reconfigurable processor that can reconfigure the connection and setting of circuit cells inside the LSI may be used.
- a part or all of the constituent elements constituting each of the above devices may be constituted by an IC card or a single module that can be attached to and detached from each device.
- the IC card or the module is a computer system including a microprocessor, a ROM, a RAM, and the like.
- the IC card or the module may include the super multifunctional LSI.
- the IC card or the module achieves its function by the microprocessor operating according to the computer program. This IC card or this module may have tamper resistance.
- the present invention may be the method described above. Further, the present invention may be a computer program that realizes these methods by a computer, or may be a digital signal composed of the computer program.
- the present invention also provides a computer-readable recording medium such as a flexible disk, hard disk, CD-ROM, MO, DVD, DVD-ROM, DVD-RAM, BD (Blu-ray Disc). ), Recorded in a semiconductor memory or the like. Further, the digital signal may be recorded on these recording media.
- a computer-readable recording medium such as a flexible disk, hard disk, CD-ROM, MO, DVD, DVD-ROM, DVD-RAM, BD (Blu-ray Disc).
- the computer program or the digital signal may be transmitted via an electric communication line, a wireless or wired communication line, a network represented by the Internet, a data broadcast, or the like.
- the present invention may also be a computer system including a microprocessor and a memory.
- the memory may store the computer program, and the microprocessor may operate according to the computer program.
- the program or the digital signal is recorded on the recording medium and transferred, or the program or the digital signal is transferred via the network or the like, and is executed by another independent computer system. It is good.
- the image encoding apparatus when encoding a stereoscopic video from two video signals, can easily view a stereoscopic video by blurring a background area and assigning information to a focus area regardless of a stereoscopic video shooting method.
- it is useful as a digital video camera, a digital still camera, or the like.
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
- Testing, Inspecting, Measuring Of Stereoscopic Televisions And Televisions (AREA)
Abstract
Description
図4は、本発明の実施の形態1における画像符号化装置の構成を示すブロック図である。
以下、本実施の形態における第1の変形例について説明する。本実施の形態では、量子化ステップを増加することによって符号化対象ブロックの画像をぼかしたが、そのブロックの高周波数領域の周波数係数を切り捨てることによってそのブロックの画像をぼかしてもよい。
以下、本実施の形態における第2の変形例について説明する。本変形例に係る画像符号化装置は焦点を調整し、その焦点にある被写体の画像をぼかすことなく、撮影距離Lが焦点よりも長い位置にある遠景が映っている領域の画像を適切にぼかす点に特徴がある。
本実施の形態における画像符号化装置は、ピクチャを構成する各ブロックに対して検出された視差ベクトルの分布(視差分布)に基づいて撮影方法を特定する点に特徴がある。
以下、本実施の形態における第1の変形例について説明する。本実施の形態では、量子化ステップを増加することによってブロックの画像をぼかしたが、実施の形態1の変形例1と同様、ブロックの高周波数領域の周波数係数を切り捨てることによってそのブロックの画像をぼかしてもよい。
以下、本実施の形態における第2の変形例について説明する。本変形例に係る画像符号化装置は、実施の形態1の変形例2と同様、焦点を調整し、その焦点にある被写体の画像をぼかすことなく、撮影距離Lが焦点よりも長い位置にある遠景が映っている領域の画像を適切にぼかす点に特徴がある。
本実施の形態では、本発明に係る画像符号化装置を備えたシステムについて説明する。
図13は、本発明の実施の形態4における画像復号装置の構成を示すブロック図である。
図14は、本発明の実施の形態5における画像復号装置の構成を示すブロック図である。
11 視差検出部
12 方法特定部
13 調整部
14 符号化部
302 輻輳角調整部
303,503,602,702 符号化調整部
304 セレクタ
305 検出部
306 減算器
307,314,1008,1009 スイッチ
308 変換部
309 量子化部
310 可変長符号化部
311 逆量子化部
312 逆変換部
313,1004 加算器
316 補償部
402 焦点調整部
601 視差分布識別部
800 画像符号化システム
801 画像符号化部
804 表示部
805 外部記録装置
1000,1100 画像復号装置
1001 可変長復号部
1002 逆量子化/逆変換部
1003 動き/視差補償部
1005,1010 デブロッキングフィルタ
1006,1011 フィルタ処理識別部
1007 エッジ強調部
Ca,Cb カメラ
D1 表示部
Me,Me1 記憶部
Claims (17)
- 立体視するための2つの画像を符号化する画像符号化装置であって、
カメラによる前記2つの画像の撮影方法を特定する方法特定部と、
前記2つの画像間の視差を、前記2つの画像のそれぞれを構成する領域ごとに検出する視差検出部と、
前記領域ごとに、前記方法特定部によって特定された撮影方法、および前記視差検出部によって当該領域に対して検出された視差に基づいて、画像をぼかすための調整量を決定する調整部と、
前記領域ごとに、当該領域に対して決定された調整量だけ当該領域の画像がぼけるように、前記2つの画像を符号化する符号化部と
を備える画像符号化装置。 - 前記方法特定部は、前記撮影方法が平行法であるか交差法であるかを特定し、
前記調整部は、
前記方法特定部によって前記撮影方法が交差法であると特定された場合には、
前記視差検出部によって検出された視差が第1の閾値より大きい領域に対して前記調整量を決定する、
請求項1に記載の画像符号化装置。 - 前記調整部は、前記視差が大きいほど、ぼかしの度合いが大きくなるような調整量を決定する、
請求項2に記載の画像符号化装置。 - 前記画像符号化装置は、さらに、
前記カメラによる撮影によって得られる前記2つの画像のそれぞれのうち、焦点が当てられている領域を特定する焦点特定部を備え、
前記調整部は、前記焦点特定部によって特定された領域に対して前記視差検出部によって検出された視差の大きさを、前記第1の閾値として設定する、
請求項2または3に記載の画像符号化装置。 - 前記方法特定部は、前記撮影方法が平行法であるか交差法であるかを特定し、
前記調整部は、
前記方法特定部によって前記撮影方法が平行法であると特定された場合には、
前記視差検出部によって検出された視差が第2の閾値より小さい領域に対して前記調整量を決定する、
請求項1に記載の画像符号化装置。 - 前記調整部は、前記視差が小さいほど、ぼかしの度合いが大きくなるような調整量を決定する、
請求項5に記載の画像符号化装置。 - 前記画像符号化装置は、さらに、
前記カメラによる撮影によって得られる前記2つの画像のそれぞれのうち、焦点が当てられている領域を特定する焦点特定部を備え、
前記調整部は、前記焦点特定部によって特定された領域に対して前記視差検出部によって検出された視差の大きさを、前記第2の閾値として設定する、
請求項5または6に記載の画像符号化装置。 - 前記方法特定部は、
前記2つの画像のそれぞれを生成するための前記カメラの2つの光軸が交わる角度を輻輳角として特定し、前記輻輳角が予め定められた角度より小さいか否かを判別し、前記輻輳角が小さくないと判別したときには、前記撮影方法が交差法であると特定し、前記輻輳角が小さいと判別したときには、前記撮影方法が平行法であると特定する、
請求項1~5の何れか1項に記載の画像符号化装置。 - 前記視差検出部は、前記2つの画像の間の視差を視差ベクトルとして前記領域ごとに検出し、
前記方法特定部は、
前記視差検出部によって領域ごとに検出される視差ベクトルの分布の中で、負の向きの視差ベクトルと正の向きの視差ベクトルとが混在しているか否かを判別し、混在していると判別したときには、前記撮影方法が交差法であると特定し、混在していないと判別したときには、前記撮影方法が平行法であると特定する、
請求項1~5の何れか1項に記載の画像符号化装置。 - 前記符号化部は、
前記2つの画像のそれぞれについて、前記領域ごとに直交変換する変換部と、
前記領域ごとに、前記変換部による直交変換によって得られた周波数係数を量子化する量子化部とを備え、
前記量子化部は、何れかの領域の周波数係数を量子化するときには、当該領域に対して前記調整部によって決定された調整量だけ量子化ステップを増加し、増加された量子化ステップを用いて当該領域の周波数係数を量子化することにより、当該領域の画像をぼかす
請求項1~9の何れか1項に記載の画像符号化装置。 - 前記符号化部は、
前記2つの画像のそれぞれについて、前記領域ごとに直交変換する変換部を備え、
前記変換部は、何れかの領域に対して直交変換を行い、当該領域に対して周波数係数を生成したときには、当該領域に含まれる周波数係数のうち、当該領域に対して前記調整部によって決定された調整量に応じた広さの高周波数領域の周波数係数を切り捨てることにより、当該領域の画像をぼかす、
請求項1~9の何れか1項に記載の画像符号化装置。 - 前記符号化部は、
前記領域ごとに検出された視差を用いて、前記2つの画像のうちの一方の画像を、他方の画像から予測することによって予測画像を生成し、前記一方の画像と前記予測画像との差分を算出することによって、前記一方の画像を符号化する、
請求項1~11の何れか1項に記載の画像符号化装置。 - 前記方法特定部は、前記2つの画像のそれぞれの撮像の向きを示す線の間の角度を前記撮影方法として特定し、
前記調整部は、前記領域ごとに、前記方法特定部によって特定された角度、および前記視差検出部によって当該領域に対して検出された視差に基づいて、画像をぼかすための調整量を決定する
請求項1に記載の画像符号化装置。 - カメラと、
前記カメラによる撮影によって生成される、立体視するための2つの画像を符号化する請求項1~13の何れか1項に記載の画像符号化装置と、
前記画像符号化装置による符号化によって生成される信号を記録するための記録媒体と
を備える画像符号化システム。 - 立体視するための2つの画像を符号化する画像符号化方法であって、
カメラによる前記2つの画像の撮影方法を特定し、
前記2つの画像間の視差を、前記2つの画像のそれぞれを構成する領域ごとに検出し、
前記領域ごとに、特定された前記撮影方法、および当該領域に対して検出された視差に基づいて、画像をぼかすための調整量を決定し、
前記領域ごとに、当該領域に対して決定された調整量だけ当該領域の画像がぼけるように、前記2つの画像を符号化する
画像符号化方法。 - 立体視するための2つの画像を符号化するためのプログラムであって、
カメラによる前記2つの画像の撮影方法を特定し、
前記2つの画像間の視差を、前記2つの画像のそれぞれを構成する領域ごとに検出し、
前記領域ごとに、特定された前記撮影方法、および当該領域に対して検出された視差に基づいて、画像をぼかすための調整量を決定し、
前記領域ごとに、当該領域に対して決定された調整量だけ当該領域の画像がぼけるように、前記2つの画像を符号化する
ことをコンピュータに実行させるプログラム。 - 立体視するための2つの画像を符号化する集積回路であって、
カメラによる前記2つの画像の撮影方法を特定する方法特定部と、
前記2つの画像間の視差を、前記2つの画像のそれぞれを構成する領域ごとに検出する視差検出部と、
前記領域ごとに、前記方法特定部によって特定された撮影方法、および前記視差検出部によって当該領域に対して検出された視差に基づいて、画像をぼかすための調整量を決定する調整部と、
前記領域ごとに、当該領域に対して決定された調整量だけ当該領域の画像がぼけるように、前記2つの画像を符号化する符号化部と
を備える集積回路。
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201180002954.0A CN102474647B (zh) | 2010-05-25 | 2011-03-30 | 图像编码装置、图像编码方法以及集成电路 |
US13/386,069 US8994788B2 (en) | 2010-05-25 | 2011-03-30 | Image coding apparatus, method, program, and circuit using blurred images based on disparity |
JP2011536224A JP5450643B2 (ja) | 2010-05-25 | 2011-03-30 | 画像符号化装置、画像符号化方法、プログラムおよび集積回路 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2010-119908 | 2010-05-25 | ||
JP2010119908 | 2010-05-25 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2011148556A1 true WO2011148556A1 (ja) | 2011-12-01 |
Family
ID=45003558
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2011/001926 WO2011148556A1 (ja) | 2010-05-25 | 2011-03-30 | 画像符号化装置、画像符号化方法、プログラムおよび集積回路 |
Country Status (4)
Country | Link |
---|---|
US (1) | US8994788B2 (ja) |
JP (1) | JP5450643B2 (ja) |
CN (1) | CN102474647B (ja) |
WO (1) | WO2011148556A1 (ja) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103220540A (zh) * | 2012-01-21 | 2013-07-24 | 瑞昱半导体股份有限公司 | 图像处理装置与方法 |
CN103369333A (zh) * | 2012-04-05 | 2013-10-23 | 联发科技(新加坡)私人有限公司 | 识别成对立体图像实际观看顺序的方法和电子装置 |
Families Citing this family (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9045961B2 (en) | 2011-01-31 | 2015-06-02 | National Oilwell Varco, L.P. | Blowout preventer seal and method of using same |
WO2013155191A2 (en) | 2012-04-10 | 2013-10-17 | National Oilwell Varco, L.P. | Blowout preventer seal assembly and method of using same |
KR20140043264A (ko) * | 2012-09-26 | 2014-04-09 | 삼성전자주식회사 | 다시점 영상 처리 장치 및 방법 |
CN102883173A (zh) * | 2012-09-27 | 2013-01-16 | 彩虹集团公司 | 一种用于3d显示的画质优化方法 |
WO2016086379A1 (en) * | 2014-12-04 | 2016-06-09 | SZ DJI Technology Co., Ltd. | Imaging system and method |
CN105100772B (zh) * | 2015-07-16 | 2017-03-15 | 深圳市华星光电技术有限公司 | 一种三维图像处理方法及装置 |
WO2020045050A1 (ja) * | 2018-08-27 | 2020-03-05 | パナソニック インテレクチュアル プロパティ コーポレーション オブ アメリカ | 符号化装置、復号装置、符号化方法及び復号方法 |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH06284449A (ja) * | 1993-03-25 | 1994-10-07 | Kyocera Corp | 視差を利用した映像信号の帯域圧縮回路および画像符号化装置 |
JPH09181966A (ja) * | 1995-12-22 | 1997-07-11 | Olympus Optical Co Ltd | 画像処理方法及び装置 |
JP2007124278A (ja) * | 2005-10-28 | 2007-05-17 | Nikon Corp | 撮像装置 |
JP2010114577A (ja) * | 2008-11-05 | 2010-05-20 | Fujifilm Corp | 撮像装置、画像処理装置、撮像装置の制御方法及び画像処理方法 |
Family Cites Families (36)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6028608A (en) * | 1997-05-09 | 2000-02-22 | Jenkins; Barry | System and method of perception-based image generation and encoding |
US6762794B1 (en) * | 1997-12-03 | 2004-07-13 | Canon Kabushiki Kaisha | Image pick-up apparatus for stereoscope |
JP2000115782A (ja) * | 1998-10-06 | 2000-04-21 | Canon Inc | 符号化装置及び方法及び記憶媒体 |
JP2000207549A (ja) * | 1999-01-11 | 2000-07-28 | Olympus Optical Co Ltd | 画像処理装置 |
JP2000251090A (ja) * | 1999-03-01 | 2000-09-14 | Sony Computer Entertainment Inc | 描画装置及び該描画装置で被写界深度を表現する方法 |
US20020009137A1 (en) * | 2000-02-01 | 2002-01-24 | Nelson John E. | Three-dimensional video broadcasting system |
US6618054B2 (en) * | 2000-05-16 | 2003-09-09 | Sun Microsystems, Inc. | Dynamic depth-of-field emulation based on eye-tracking |
US6847735B2 (en) * | 2000-06-07 | 2005-01-25 | Canon Kabushiki Kaisha | Image processing system, image processing apparatus, image input apparatus, image output apparatus and method, and storage medium |
TW527518B (en) * | 2000-07-14 | 2003-04-11 | Massachusetts Inst Technology | Method and system for high resolution, ultra fast, 3-D imaging |
JP3659157B2 (ja) * | 2000-10-17 | 2005-06-15 | 日本電気株式会社 | 映像内容に重み付けをする画像圧縮方式 |
US20030151689A1 (en) * | 2002-02-11 | 2003-08-14 | Murphy Charles Douglas | Digital images with composite exposure |
WO2004077820A1 (ja) * | 2003-02-25 | 2004-09-10 | Matsushita Electric Industrial Co. Ltd. | 画像撮像処理方法および画像撮像装置 |
US7532234B2 (en) * | 2003-06-19 | 2009-05-12 | Microsoft Corporation | Automatic analysis and adjustment of digital images upon acquisition |
US20050017968A1 (en) * | 2003-07-21 | 2005-01-27 | Stephan Wurmlin | Differential stream of point samples for real-time 3D video |
WO2006074310A2 (en) * | 2005-01-07 | 2006-07-13 | Gesturetek, Inc. | Creating 3d images of objects by illuminating with infrared patterns |
JP4630149B2 (ja) | 2005-07-26 | 2011-02-09 | シャープ株式会社 | 画像処理装置 |
US7916934B2 (en) * | 2006-04-04 | 2011-03-29 | Mitsubishi Electric Research Laboratories, Inc. | Method and system for acquiring, encoding, decoding and displaying 3D light fields |
WO2007116549A1 (ja) * | 2006-04-07 | 2007-10-18 | Sharp Kabushiki Kaisha | 画像処理装置 |
JP4757812B2 (ja) * | 2007-02-20 | 2011-08-24 | 富士フイルム株式会社 | 立体撮影装置および方法並びにプログラム |
US8594180B2 (en) * | 2007-02-21 | 2013-11-26 | Qualcomm Incorporated | 3D video encoding |
WO2008140656A2 (en) * | 2007-04-03 | 2008-11-20 | Gary Demos | Flowfield motion compensation for video compression |
US8031954B2 (en) * | 2007-04-26 | 2011-10-04 | Canon Kabushiki Kaisha | Image encoding apparatus and control method thereof using prediction encoding and pixel classification |
JP5040440B2 (ja) * | 2007-05-23 | 2012-10-03 | ソニー株式会社 | 撮像防止装置、撮像防止方法 |
JP5109803B2 (ja) * | 2007-06-06 | 2012-12-26 | ソニー株式会社 | 画像処理装置、画像処理方法及び画像処理プログラム |
FR2917845B1 (fr) * | 2007-06-19 | 2011-08-19 | Christophe Brossier | Procede de visualisation d'une sequence d'images produisant une sensation de relief |
US7817187B2 (en) * | 2007-06-27 | 2010-10-19 | Aptina Imaging Corporation | Image blur correction using a secondary camera |
US7852980B2 (en) * | 2007-10-31 | 2010-12-14 | Xoran Technologies, Inc. | Computed tomography with resolution recovery |
US8107700B2 (en) * | 2007-11-21 | 2012-01-31 | Merge Cad Inc. | System and method for efficient workflow in reading medical image data |
KR101520619B1 (ko) * | 2008-02-20 | 2015-05-18 | 삼성전자주식회사 | 스테레오 동기화를 위한 스테레오스코픽 영상의 시점 결정방법 및 장치 |
US8405708B2 (en) * | 2008-06-06 | 2013-03-26 | Reald Inc. | Blur enhancement of stereoscopic images |
US8184196B2 (en) * | 2008-08-05 | 2012-05-22 | Qualcomm Incorporated | System and method to generate depth data using edge detection |
KR101520620B1 (ko) * | 2008-08-18 | 2015-05-18 | 삼성전자주식회사 | 2차원/3차원 재생 모드 결정 방법 및 장치 |
JP4625517B2 (ja) * | 2008-10-27 | 2011-02-02 | 富士フイルム株式会社 | 3次元表示装置および方法並びにプログラム |
US20110304689A1 (en) * | 2010-06-15 | 2011-12-15 | Sharp Laboratories Of America, Inc. | Remote user control for stereoscopic display |
US9123115B2 (en) * | 2010-11-23 | 2015-09-01 | Qualcomm Incorporated | Depth estimation based on global motion and optical flow |
US9076267B2 (en) * | 2011-07-19 | 2015-07-07 | Panasonic Intellectual Property Corporation Of America | Image coding device, integrated circuit thereof, and image coding method |
-
2011
- 2011-03-30 WO PCT/JP2011/001926 patent/WO2011148556A1/ja active Application Filing
- 2011-03-30 US US13/386,069 patent/US8994788B2/en not_active Expired - Fee Related
- 2011-03-30 CN CN201180002954.0A patent/CN102474647B/zh not_active Expired - Fee Related
- 2011-03-30 JP JP2011536224A patent/JP5450643B2/ja not_active Expired - Fee Related
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH06284449A (ja) * | 1993-03-25 | 1994-10-07 | Kyocera Corp | 視差を利用した映像信号の帯域圧縮回路および画像符号化装置 |
JPH09181966A (ja) * | 1995-12-22 | 1997-07-11 | Olympus Optical Co Ltd | 画像処理方法及び装置 |
JP2007124278A (ja) * | 2005-10-28 | 2007-05-17 | Nikon Corp | 撮像装置 |
JP2010114577A (ja) * | 2008-11-05 | 2010-05-20 | Fujifilm Corp | 撮像装置、画像処理装置、撮像装置の制御方法及び画像処理方法 |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103220540A (zh) * | 2012-01-21 | 2013-07-24 | 瑞昱半导体股份有限公司 | 图像处理装置与方法 |
CN103220540B (zh) * | 2012-01-21 | 2017-04-12 | 瑞昱半导体股份有限公司 | 图像处理装置与方法 |
CN103369333A (zh) * | 2012-04-05 | 2013-10-23 | 联发科技(新加坡)私人有限公司 | 识别成对立体图像实际观看顺序的方法和电子装置 |
US9031316B2 (en) | 2012-04-05 | 2015-05-12 | Mediatek Singapore Pte. Ltd. | Method for identifying view order of image frames of stereo image pair according to image characteristics and related machine readable medium thereof |
CN103369333B (zh) * | 2012-04-05 | 2015-11-18 | 联发科技(新加坡)私人有限公司 | 识别成对立体图像实际观看顺序的方法和电子装置 |
Also Published As
Publication number | Publication date |
---|---|
CN102474647A (zh) | 2012-05-23 |
JPWO2011148556A1 (ja) | 2013-07-25 |
US8994788B2 (en) | 2015-03-31 |
JP5450643B2 (ja) | 2014-03-26 |
US20120120193A1 (en) | 2012-05-17 |
CN102474647B (zh) | 2015-08-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP5450643B2 (ja) | 画像符号化装置、画像符号化方法、プログラムおよび集積回路 | |
JP6178017B2 (ja) | ステレオビデオのための深度認識向上 | |
JP5565001B2 (ja) | 立体映像撮像装置、立体映像処理装置および立体映像撮像方法 | |
WO2013011608A1 (ja) | 画像符号化装置とその集積回路、および画像符号化方法 | |
CN106851239B (zh) | 用于使用视差信息的3d媒体数据产生、编码、解码和显示的方法和装置 | |
WO2012111757A1 (ja) | 画像処理装置および画像処理方法 | |
US20180262754A1 (en) | Image compression apparatus, image decoding apparatus, and image processing method | |
JP2012257198A (ja) | 立体画像符号化装置、その方法、および立体画像符号化装置を有する撮像装置 | |
WO2012060156A1 (ja) | 多視点画像符号化装置及び多視点画像復号装置 | |
JP4964355B2 (ja) | 立体映像符号化装置、立体映像撮影装置、および立体映像符号化方法 | |
JP5395911B2 (ja) | ステレオ画像符号化装置、方法 | |
WO2011074189A1 (ja) | 画像符号化方法及び画像符号化装置 | |
JP2013051637A (ja) | 符号化装置及び符号化方法 | |
JP5869839B2 (ja) | 画像処理装置およびその制御方法 | |
JP6341598B2 (ja) | 画像符号化装置、画像復号装置、画像符号化プログラム及び画像復号プログラム | |
JP4375662B2 (ja) | 画像処理装置、画像処理方法、プログラム、情報記録媒体及び撮像装置 | |
JP2012178818A (ja) | 映像符号化装置および映像符号化方法 | |
Liu et al. | Motion blur compensation in HEVC using fixed-length adaptive filter | |
Phi | Perceptually Optimized Plenoptic Data Representation and Coding | |
JP4136680B2 (ja) | 撮像装置、撮像方法、コンピュータプログラム及び記録媒体 | |
WO2011102131A1 (ja) | 画像符号化装置、画像符号化方法、プログラム、及び集積回路 | |
JP2005143029A (ja) | 画像処理装置、画像処理方法、プログラム、情報記録媒体及び撮像装置 | |
BR112016020544B1 (pt) | Realce consciente de profundidade para vídeo estéreo | |
JP2013085064A (ja) | 多視点画像符号化装置、多視点画像復号装置、多視点画像符号化方法及び多視点画像復号方法 | |
JP2012147073A (ja) | 画像符号化装置、画像符号化方法、及び撮像システム |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 201180002954.0 Country of ref document: CN |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2011536224 Country of ref document: JP |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 11786260 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 13386069 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 11786260 Country of ref document: EP Kind code of ref document: A1 |