CN103198473A - Depth image generating method and device - Google Patents
Depth image generating method and device Download PDFInfo
- Publication number
- CN103198473A CN103198473A CN2013100696403A CN201310069640A CN103198473A CN 103198473 A CN103198473 A CN 103198473A CN 2013100696403 A CN2013100696403 A CN 2013100696403A CN 201310069640 A CN201310069640 A CN 201310069640A CN 103198473 A CN103198473 A CN 103198473A
- Authority
- CN
- China
- Prior art keywords
- depth information
- image
- image object
- pars intermedia
- depth
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Image Processing (AREA)
- Processing Or Creating Images (AREA)
Abstract
The invention discloses a depth image generating method and a depth image generating device. The method includes the steps: conducting portioning operation to a currently input image and obtaining object collection of the image, wherein the object collection of the image comprises one or more than one of a sky image object, a ground image object and a middle portion image object; respectively selecting a depth information algorithm for each image object in the object collection of the image from preset depth information algorithms and computing depth information of a corresponding image object; and enabling depth information of each image object to be combined to obtain the depth image of the input image. By the adoption of the method and the device, the precise depth image of the input image can be quickly obtained, and particularly in the scene that certain requirements to precision of depth information are not high and only distance relations are needed, the requirements that the depth image is obtained quickly can be met, computing resources are saved, and cost is lowered.
Description
Technical field
The present invention relates to image processing techniques, relate in particular to a kind of degree of depth drawing generating method and device.
Background technology
In the image processing techniques about streetscape, an important problem is obtaining of picture depth information, under each scene for streetscape, each point can be represented with depth map (Depth Map) with respect to the distance of video camera in each image, and certain is a bit and the distance between the video camera in each the pixel value presentation video in the depth map.
At present, obtain current shooting to the technology of depth map of scene image can be divided into the passive ranging sensing and the sensing two big classes of initiatively finding range.The passive ranging sensing refers to that vision system receives the luminous energy from scene emission or reflection, forms the gray level image of relevant light energy distribution, and then obtains the depth information of scene.The sensing of initiatively finding range refers to vision system initiatively to the scene emitted energy, and by accepting the reflected energy compute depth information of scene, the most frequently used sensor-based system of initiatively finding range is radar ranging system and range of triangle system.
Above-mentioned two class modes can both be carried out depth calculation to each point in the image, then each point are carried out combination, just can obtain accurate depth map about this image.The account form of prior art is that all the object unifications in the image are calculated, and is comparatively complicated, can't finish picture depth fast and handle.
Summary of the invention
Embodiment of the invention technical matters to be solved is, a kind of degree of depth drawing generating method and device are provided, and can finish picture depth comparatively rapidly and calculate, and obtains the depth map of image.
In order to solve the problems of the technologies described above, the embodiment of the invention provides a kind of degree of depth drawing generating method, comprising:
Image to current input carries out cutting operation, obtains the object set of described image, comprises in sky image object, ground image object and the pars intermedia image object any one or more in the object set of described image;
From the depth information algorithm that presets, be respectively each the image object selected depth information algorithm in the object set of described image, calculate the depth information of correspondence image object;
With the depth information combination of each image object, obtain the depth map of the image of described input.
Wherein, if comprise the ground image object in the object set of described image, each image object selected depth information algorithm in the described object set that is respectively described image from the depth information algorithm that presets then calculates the depth information of correspondence image object, comprising:
From the depth information algorithm that presets, be the depth information algorithm of the ground image Object Selection in the object set of described image based on interpolation calculation;
Calculate the depth information of at least three points in the ground image object;
According to the depth information of at least three points that calculate, adopt the mode of interpolation calculation to obtain the depth information of having a few in the described ground image object.
Wherein, if comprise the pars intermedia image object in the object set of described image, each image object selected depth information algorithm in the described object set that is respectively described image from the depth information algorithm that presets then calculates the depth information of correspondence image object, comprising:
From the depth information algorithm that presets, be the depth information algorithm of the ground image Object Selection in the object set of described image based on the binocular solid geometric relationship;
Be base unit with unit plane in the described pars intermedia image object, employing is calculated the depth information on constituent parts plane in the described pars intermedia image object based on the depth information algorithm of binocular solid geometric relationship, and described unit plane refers to that the area size is the plane of default unit area size in the pars intermedia image object;
According to the depth information on constituent parts plane, combination obtains the depth information of described pars intermedia image object.
Wherein, described is base unit with unit plane in the described pars intermedia image object, adopts and calculates the depth information on constituent parts plane in the described pars intermedia image object based on the depth information algorithm of binocular solid geometric relationship, comprising:
Described pars intermedia image object is carried out profile extract operation;
The result who extracts operation according to profile carries out modeling to described pars intermedia image object, obtains the three-dimensional wire-frame model of pars intermedia image object;
According to default unit area each face in the three-dimensional wire-frame model of described pars intermedia image object is divided, obtained the unit plane of each face correspondence in the three-dimensional wire-frame model;
Adopt the depth calculation mode of binocular solid geometric relationship, unit in the three-dimensional wire-frame model is calculated on the plane, obtain the depth information on each unit plane.
Wherein, if comprise the sky image object in the object set of described image, the depth information that then calculates the sky image object in the object set of described image comprises:
The depth information of the sky image object in the object set of described image is labeled as the infinite distance.
Correspondingly, the embodiment of the invention also provides a kind of depth map generating apparatus, comprising:
Cut apart module, be used for the image of input is carried out cutting operation, obtain the object set of image, comprise in sky image object, ground image object and the pars intermedia image object any one or more in the object set of described image;
Computing module for each the image object selected depth information algorithm the object set that is respectively described image from the depth information algorithm that presets, calculates the depth information of correspondence image object;
Composite module is used for the depth information combination with each image object, obtains the depth map of the image of described input.
Wherein, described computing module comprises:
First selected cell is used for being the depth information algorithm of the ground image Object Selection the object set of described image based on interpolation calculation from the depth information algorithm that presets;
First computing unit is for the depth information that calculates at least three points of ground image object;
First processing unit is used for the depth information according at least three points that calculate, and adopts the mode of interpolation to obtain the depth information of having a few in the described ground image object.
Wherein, described computing module comprises:
Second selected cell is used for being the depth information algorithm of the ground image Object Selection the object set of described image based on the binocular solid geometric relationship from the depth information algorithm that presets;
Second computing unit, being used for described pars intermedia image object unit plane is base unit, employing is calculated the depth information on constituent parts plane in the described pars intermedia image object based on the depth information algorithm of binocular solid geometric relationship, and described unit plane refers to that the area size is the plane of default unit area size in the pars intermedia image object;
Second processing unit, according to the depth information on constituent parts plane, combination obtains the depth information of described pars intermedia image object.
Wherein, described second computing unit comprises:
Extract subelement, described pars intermedia image object is carried out profile extract operation;
The modeling subelement is used for extracting the result who operates according to profile described pars intermedia image object is carried out modeling, obtains the three-dimensional wire-frame model of pars intermedia image object;
Subelement is divided on the plane, is used for according to default unit area each face of the three-dimensional wire-frame model of described pars intermedia image object being divided, and obtains the unit plane of each face correspondence in the three-dimensional wire-frame model;
Computation subunit for the depth calculation mode that adopts the binocular solid geometric relationship, is calculated on the plane unit in the three-dimensional wire-frame model, obtains the depth information on each unit plane.
Wherein, described computing module comprises:
Indexing unit is used for the depth information of the sky image object of the object set of described image is labeled as the infinite distance.
Implement the embodiment of the invention, have following beneficial effect:
The embodiment of the invention obtains different image objects by the image of input is cut apart, calculate the depth information of this image object then at different image object selected depth information algorithms, the inventor finds, can obtain the depth map comparatively accurately of input picture rapidly, particularly not high to the accuracy requirement of depth information at some, and only require and can satisfy the requirement of obtaining fast of depth map in the scene of far and near relation, save computational resource, reduced cost.
Description of drawings
In order to be illustrated more clearly in the embodiment of the invention or technical scheme of the prior art, to do to introduce simply to the accompanying drawing of required use in embodiment or the description of the Prior Art below, apparently, accompanying drawing in describing below only is some embodiments of the present invention, for those of ordinary skills, under the prerequisite of not paying creative work, can also obtain other accompanying drawing according to these accompanying drawings.
Fig. 1 is the schematic flow sheet of a kind of degree of depth drawing generating method of the embodiment of the invention;
Fig. 2 is the schematic flow sheet of the method for the embodiment of the invention a kind of depth information of calculating the correspondence image object;
Fig. 3 is the schematic flow sheet of the method for the another kind of the embodiment of the invention depth information that calculates the correspondence image object;
Fig. 4 is the schematic flow sheet of the another kind of degree of depth drawing generating method of the embodiment of the invention;
Fig. 5 is that the structure of a kind of depth map generating apparatus of the embodiment of the invention is formed synoptic diagram;
Fig. 6 is wherein a kind of concrete structure synoptic diagram of computing module described in Fig. 5;
Fig. 7 is the wherein another kind of concrete structure synoptic diagram of computing module described in Fig. 5.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the invention, the technical scheme in the embodiment of the invention is clearly and completely described, obviously, described embodiment only is the present invention's part embodiment, rather than whole embodiment.Based on the embodiment among the present invention, those of ordinary skills belong to the scope of protection of the invention not making the every other embodiment that obtains under the creative work prerequisite.
Seeing also Fig. 1, is the schematic flow sheet of a kind of degree of depth drawing generating method of the embodiment of the invention, and the described method of the embodiment of the invention can be applicable in all kinds of computer vision systems, and is concrete, and the described method of present embodiment comprises:
S101: the image to current input carries out cutting operation, obtains the object set of described image, comprises in sky image object, ground image object and the pars intermedia image object any one or more in the object set of described image.
The image of current input can be under scenes such as streetscape shooting, is taken the binocular image that obtains by video camera.Generally comprise sky, ground and the center section between sky and ground in the image, center section can be images such as buildings, street trees, vehicle, sculpture.
In described S101, the sky of image partly cut apart obtain the sky image object and can realize in the following manner:
The colouring information intrinsic according to sky, and use mean shift meanshift algorithm to carry out cutting operation, obtain the sky image object.
The above ground portion of image cut apart obtains the ground image object and then can realize in the following manner:
The first step, dense Stereo Matching generates disparity map on the image; In second step, the ground point sampling can comprise that employing is based on the sample mode of subpoint density; In the 3rd step, the ground match can be adopted the Consensus based on RANSAC(RANdom SAmple, the random sampling consistance) plane fitting calculates; In the 4th step, the result of ground match is carried out just cutting apart; The 5th step, ground point ON operation and computation bound; In the 6th step, carry out ground degree of depth interpolation.Wherein, the above-mentioned first step is according to the associated information calculation disparity map in the image space, and second step to the 4th step finishes in parallax space (three dimensions), and then shines upon back and finish subsequent operation in the image space.That is to say that in fact whole ground is cut apart and comprised from image space to the parallax space and then to the cutting procedure of image space.
After the cutting apart of the sky of finishing image and ground, objects such as the buildings in the center section can be directly as the pars intermedia image object, to carry out subsequent treatment.
S102: from the depth information algorithm that presets, be respectively each the image object selected depth information algorithm in the object set of described image, calculate the depth information of correspondence image object;
The described depth information algorithm that presets can comprise multiple, according to the type of cutting apart the image object that obtains, can select different depth information algorithms, concrete, for the sky image object, directly its depth information is labeled as the infinite distance, then the sky image object is expressed as the infinite distance in depth map; The ground image object is then used the mode compute depth value of interpolation; The pars intermedia image object then carries out comparatively detailed calculating based on the depth information algorithm of binocular solid geometric relationship.
S103: with the depth information combination of each image object, obtain the depth map of the image of described input.
Because the depth map of described image is that therefore each point is recorded to the picture depth figure that can obtain described input on the depth map with the depth information that obtains to the distance of same video camera on the presentation video.
The embodiment of the invention obtains different image objects by the image of input is cut apart, calculate the depth information of this image object then at different image object selected depth information algorithms, the inventor finds, can obtain the depth map comparatively accurately of input picture rapidly, particularly not high to the accuracy requirement of depth information at some, and only require and can satisfy the requirement of obtaining fast of depth map in the scene of far and near relation, save computational resource, reduced cost.
See also Fig. 2 again, it is the schematic flow sheet of the method for the embodiment of the invention a kind of depth information of calculating the correspondence image object, the described method of the embodiment of the invention is corresponding to the S102 in above-described embodiment, concrete, the described method of the embodiment of the invention is used for when the object set of described image comprises the ground image object, to the computing method of the depth information of ground image object, it comprises:
S201: from the depth information algorithm that presets, be the depth information algorithm of the ground image Object Selection in the object set of described image based on interpolation calculation;
S202: the depth information that calculates at least three points in the ground image object;
S203: according to the depth information of at least three points that calculate, adopt the mode of interpolation calculation to obtain the depth information of having a few in the described ground image object.
Concrete, depth information algorithm based on interpolation calculation is according to selecting suitable specific function, each aspect in the interval of ground image object calculates the part point value, use the approximate value of other points of value of the specific function of selecting then at other aspects in interval, the embodiment of the invention specifically can adopt piece commonly used at present to mate the depth information that (Block Match) calculates part point, obtains the depth information of other points then by interpolation.After obtaining the depth information of each point, namely finish the calculating of depth information in the described ground image object, finish the depth map of ground image object part.
The embodiment of the invention adopts the mode of interpolation to calculate the depth information of above ground portion, after calculating several points, just can obtain the depth value of other points, satisfied the requirement of quick compute depth information, and can more efficiently get rid of such as the influence to the above ground portion depth information such as trees, vehicle.
See also Fig. 3 again, it is the schematic flow sheet of the method for the another kind of the embodiment of the invention depth information that calculates the correspondence image object, the described method of the embodiment of the invention is corresponding to the S102 in above-described embodiment, concrete, the described method of the embodiment of the invention is used for when the object set of described image comprises the pars intermedia image object, computing method to the depth information of pars intermedia image object, the actual image object that can comprise objects such as buildings, personage, vehicle of pars intermedia image object, the described method of present embodiment comprises:
S301: from the depth information algorithm that presets, be the depth information algorithm of the ground image Object Selection in the object set of described image based on the binocular solid geometric relationship;
S302: be base unit with unit plane in the described pars intermedia image object, employing is calculated the depth information on constituent parts plane in the described pars intermedia image object based on the depth information algorithm of binocular solid geometric relationship, and described unit plane refers to that the area size is the plane of default unit area size in the pars intermedia image object;
S303: according to the depth information on constituent parts plane, combination obtains the depth information of described pars intermedia image object.
Depth information algorithm based on the binocular solid geometric relationship is the depth information that obtains impact point by the distance of same impact point to two an identical video camera on the image object, in the embodiment of the invention, be base unit with unit plane in the described pars intermedia image object, replace impact point with the point set that constitutes of being had a few in the unit plane, to calculate the depth information on this unit plane.
Image for input, the imaging plane that can suppose image is positioned on the same plane, and the coordinate axis of two video cameras is parallel to each other, and the x axle overlaps, the distance of video camera on the x direction of principal axis is called parallax range, and the alternate position spike of impact point is called parallax in the image that obtains separately in two video cameras.Concern similarity, parallax range, parallax and two focus of camera of intermediate cam shape then according to solid geometry, can obtain the above impact point of image is the depth information of point set, thereby obtains the depth information on this unit plane.
Further, described S302 specifically can comprise:
Described pars intermedia image object is carried out profile extract operation;
The result who extracts operation according to profile carries out modeling to described pars intermedia image object, obtains the three-dimensional wire-frame model of pars intermedia image object;
According to default unit area each face in the three-dimensional wire-frame model of described pars intermedia image object is divided, obtained the unit plane of each face correspondence in the three-dimensional wire-frame model;
Adopt the depth calculation mode of binocular solid geometric relationship, unit in the three-dimensional wire-frame model is calculated on the plane, obtain the depth information on each unit plane.
Be the modeling of described pars intermedia image object according to profile at first namely, for example, adopt right cylinder to represent trees, people according to profile, adopt rectangular parallelepiped, cube to represent building, the geometry that like this can be further each object of pars intermedia be described as rule roughly is wire-frame model, and then be the depth information that base unit calculates each point set (unit plane) in the wire-frame model with the point set, this can improve the speed of depth calculation further, and saving assesses the cost.
In the embodiment of the invention by being the depth information that base unit calculates objects such as buildings, people, vehicle based on the depth information algorithm of binocular solid geometric relationship, the unit plane formed with point set, need not carry out the calculating of each point, the computing velocity of depth information can be provided significantly, and can also improve the speed of depth calculation further by the mode of modeling, saving assesses the cost.
Further, seeing also Fig. 4 again, is the schematic flow sheet of the another kind of degree of depth drawing generating method of the embodiment of the invention, and the described method of the embodiment of the invention can be applicable in all kinds of computer vision systems, and is concrete, and the described method of present embodiment comprises:
S401: input picture;
S402: the image to current input carries out cutting operation, cuts apart and obtains the sky image object;
S403: the image to current input carries out cutting operation, cuts apart to obtain ground image object and pars intermedia image object;
S404: calculate the depth information of having a few in the described ground image object;
Described S404 specifically can comprise: be the depth information algorithm of the ground image Object Selection in the object set of described image based on interpolation calculation from the depth information algorithm that presets; Calculate the depth information of at least three points in the ground image object; According to the depth information of at least three points that calculate, adopt the mode of interpolation calculation to obtain the depth information of having a few in the described ground image object.
S405: the depth information that calculates described pars intermedia image object;
Described S405 specifically can comprise: be the depth information algorithm of the ground image Object Selection in the object set of described image based on the binocular solid geometric relationship from the depth information algorithm that presets; Be base unit with unit plane in the described pars intermedia image object, employing is calculated the depth information on constituent parts plane in the described pars intermedia image object based on the depth information algorithm of binocular solid geometric relationship, and described unit plane refers to that the area size is the plane of default unit area size in the pars intermedia image object; According to the depth information on constituent parts plane, combination obtains the depth information of described pars intermedia image object.
Wherein, be base unit with unit plane in the described pars intermedia image object, the depth information that adopts depth information algorithm based on the binocular solid geometric relationship to calculate constituent parts plane in the described pars intermedia image object comprises: described pars intermedia image object is carried out profile extract operation; The result who extracts operation according to profile carries out modeling to described pars intermedia image object, obtains the three-dimensional wire-frame model of pars intermedia image object; According to default unit area each face in the three-dimensional wire-frame model of described pars intermedia image object is divided, obtained the unit plane of each face correspondence in the three-dimensional wire-frame model; Adopt the depth calculation mode of binocular solid geometric relationship, unit in the three-dimensional wire-frame model is calculated on the plane, obtain the depth information on each unit plane.
S406: the depth information of the sky image object in the object set of described image is labeled as the infinite distance;
S407: with the depth information combination of each image object, obtain the depth map of the image of described input.
The embodiment of the invention obtains different image objects by the image of input is cut apart, calculate the depth information of this image object then at different image object selected depth information algorithms, the inventor finds, can obtain the depth map comparatively accurately of input picture rapidly, particularly not high to the accuracy requirement of depth information at some, and only require and can satisfy the requirement of obtaining fast of depth map in the scene of far and near relation, save computational resource, reduced cost.
Depth map generating apparatus to the embodiment of the invention is described in detail below.
Seeing also Fig. 5, is that the structure of a kind of depth map generating apparatus of the embodiment of the invention is formed synoptic diagram, and the described device of the embodiment of the invention can be applicable in all kinds of computer vision systems, and is concrete, and the described device of present embodiment comprises:
Cut apart module 1, be used for the image of input is carried out cutting operation, obtain the object set of image, comprise in sky image object, ground image object and the pars intermedia image object any one or more in the object set of described image;
The image of current input can be under scenes such as streetscape shooting, is taken the binocular image that obtains by video camera.Generally comprise sky, ground and the center section between sky and ground in the image, center section can be images such as buildings, street trees, vehicle, sculpture.
The described cutting operation of cutting apart the image of 1 pair of input of module specifically can not repeat them here referring to the partitioning scheme among the method embodiment of Fig. 1 correspondence.
The described depth information algorithm that presets can comprise multiple, described computing module 2 specifically can be according to the type of cutting apart the image object that obtains, select different depth information algorithms namely to calculate, concrete, described computing module 2 is for the sky image object, directly its depth information is labeled as the infinite distance, then the sky image object is expressed as the infinite distance in depth map; Then use the mode compute depth value of interpolation for the ground image object; Then carry out comparatively detailed calculating based on the depth information algorithm of binocular solid geometric relationship for the pars intermedia image object.
Because the depth map of described image is that therefore each point is recorded to the picture depth figure that can obtain described input on the depth map with the depth information that obtains to the distance of same video camera on the presentation video.
The embodiment of the invention is identical to the partitioning scheme that the image of input carries out described in the corresponding embodiment of cutting apart of cutting operation of above-mentioned and above-mentioned Fig. 1, does not give unnecessary details at this.
The embodiment of the invention obtains different image objects by the image of input is cut apart, calculate the depth information of this image object then at different image object selected depth information algorithms, the inventor finds, can obtain the depth map comparatively accurately of input picture rapidly, particularly not high to the accuracy requirement of depth information at some, and only require and can satisfy the requirement of obtaining fast of depth map in the scene of far and near relation, save computational resource, reduced cost.
Further, seeing also Fig. 6, is wherein a kind of concrete structure synoptic diagram of computing module described in Fig. 5, and described computing module 2 specifically carries out depth calculation by each unit among Fig. 6 to the ground image object, and concrete described computing module 2 comprises:
First selected cell 21 is used for being the depth information algorithm of the ground image Object Selection the object set of described image based on interpolation calculation from the depth information algorithm that presets;
First computing unit 22 is for the depth information that calculates at least three points of ground image object;
First processing unit 23 is used for the depth information according at least three points that calculate, and adopts the mode of interpolation to obtain the depth information of having a few in the described ground image object.
Concrete, depth information algorithm based on interpolation calculation is according to selecting suitable specific function, each aspect in the interval of ground image object calculates the part point value, use the approximate value of other points of value of the specific function of selecting then at other aspects in interval, described computing module 2 is by described first computing unit 22 and first processing unit 23, adopt piece commonly used at present to mate the depth information that (Block Match) calculates part point, obtain the depth information of other points then by interpolation.After obtaining the depth information of each point, namely finish the calculating of depth information in the described ground image object, finish the depth map of ground image object part.
Described computing module 2 adopts the mode of interpolation to calculate the depth information of above ground portion, after calculating several points, just can obtain the depth value of other points, satisfied the requirement of quick compute depth information, and can more efficiently get rid of such as the influence to the above ground portion depth information such as trees, vehicle.
Further, seeing also Fig. 7, is the wherein another kind of concrete structure synoptic diagram of computing module described in Fig. 5, and described computing module 2 specifically carries out depth calculation by portion's image object between each cell pairs among Fig. 7, and concrete described computing module 2 comprises:
Second selected cell 24 is used for being the depth information algorithm of the ground image Object Selection the object set of described image based on the binocular solid geometric relationship from the depth information algorithm that presets;
Depth information algorithm based on the binocular solid geometric relationship is the depth information that obtains impact point by the distance of same impact point to two an identical video camera on the image object, described computing module 2 is base unit by described second computing unit 25 and second processing unit 26 with unit plane in the described pars intermedia image object, replace impact point with the point set that constitutes of being had a few in the unit plane, calculate the depth information on this unit plane.
By described second computing unit 25 and second processing unit 26, image for input, the imaging plane that can suppose image is positioned on the same plane, the coordinate axis of two video cameras is parallel to each other, and the x axle overlaps, the distance of video camera on the x direction of principal axis is called parallax range, and the alternate position spike of impact point is called parallax in the image that obtains separately in two video cameras.Concern similarity, parallax range, parallax and two focus of camera of intermediate cam shape then according to solid geometry, can obtain the above impact point of image is the depth information of point set, thereby obtains the depth information on this unit plane.
Further concrete, described second computing unit 24 can also comprise that following subelement finishes corresponding function:
Extract subelement, described pars intermedia image object is carried out profile extract operation;
The modeling subelement is used for extracting the result who operates according to profile described pars intermedia image object is carried out modeling, obtains the three-dimensional wire-frame model of pars intermedia image object;
Subelement is divided on the plane, is used for according to default unit area each face of the three-dimensional wire-frame model of described pars intermedia image object being divided, and obtains the unit plane of each face correspondence in the three-dimensional wire-frame model;
Computation subunit for the depth calculation mode that adopts the binocular solid geometric relationship, is calculated on the plane unit in the three-dimensional wire-frame model, obtains the depth information on each unit plane.
Described second computing unit 24 is by above-mentioned subelement, be the modeling of described pars intermedia image object according to profile at first, for example, adopt right cylinder to represent trees, people according to profile, adopt rectangular parallelepiped, cube to represent building, the geometry that like this can be further each object of pars intermedia be described as rule roughly is wire-frame model, and then be the depth information that base unit calculates each point set (unit plane) in the wire-frame model with the point set, this can improve the speed of depth calculation further, and saving assesses the cost.
Described computing module 2 is by being the depth information that base unit calculates objects such as buildings, people, vehicle based on the depth information algorithm of binocular solid geometric relationship, the unit plane formed with point set, need not carry out the calculating of each point, the computing velocity of depth information can be provided significantly, and can also improve the speed of depth calculation further by the mode of modeling, saving assesses the cost.
Further, for the sky image object, described computing module 2 can comprise that indexing unit handles: indexing unit is used for the depth information of the sky image object of the object set of described image is labeled as the infinite distance.
Need to prove, described computing module 2 can comprise first selected cell 21, first computing unit 22, first processing unit 23 simultaneously, and second selected cell 24, second computing unit 25, second processing unit 26, and indexing unit, certainly, also can obtain scene according to real image, the corresponding function of computing module 2 is finished in selection part unit wherein.
The embodiment of the invention can obtain the depth map comparatively accurately of input picture rapidly, particularly not high to the accuracy requirement of depth information at some, and only require and can satisfy the requirement of obtaining fast of depth map in the scene of far and near relation, save computational resource, reduced cost.
One of ordinary skill in the art will appreciate that all or part of flow process that realizes in above-described embodiment method, be to instruct relevant hardware to finish by computer program, described program can be stored in the computer read/write memory medium, this program can comprise the flow process as the embodiment of above-mentioned each side method when carrying out.Wherein, described storage medium can be magnetic disc, CD, read-only storage memory body (Read-Only Memory, ROM) or at random store memory body (Random Access Memory, RAM) etc.
Above disclosed is preferred embodiment of the present invention only, can not limit the present invention's interest field certainly with this, and therefore the equivalent variations of doing according to claim of the present invention still belongs to the scope that the present invention is contained.
Claims (10)
1. a degree of depth drawing generating method is characterized in that, comprising:
Image to current input carries out cutting operation, obtains the object set of described image, comprises in sky image object, ground image object and the pars intermedia image object any one or more in the object set of described image;
From the depth information algorithm that presets, be respectively each the image object selected depth information algorithm in the object set of described image, calculate the depth information of correspondence image object;
With the depth information combination of each image object, obtain the depth map of the image of described input.
2. the method for claim 1, it is characterized in that, if comprise the ground image object in the object set of described image, each image object selected depth information algorithm in the described object set that from the depth information algorithm that presets, is respectively described image then, calculate the depth information of correspondence image object, comprising:
From the depth information algorithm that presets, be the depth information algorithm of the ground image Object Selection in the object set of described image based on interpolation calculation;
Calculate the depth information of at least three points in the ground image object;
According to the depth information of at least three points that calculate, adopt the mode of interpolation calculation to obtain the depth information of having a few in the described ground image object.
3. the method for claim 1, it is characterized in that, if comprise the pars intermedia image object in the object set of described image, each image object selected depth information algorithm in the described object set that from the depth information algorithm that presets, is respectively described image then, calculate the depth information of correspondence image object, comprising:
From the depth information algorithm that presets, be the depth information algorithm of the ground image Object Selection in the object set of described image based on the binocular solid geometric relationship;
Be base unit with unit plane in the described pars intermedia image object, employing is calculated the depth information on constituent parts plane in the described pars intermedia image object based on the depth information algorithm of binocular solid geometric relationship, and described unit plane refers to that the area size is the plane of default unit area size in the pars intermedia image object;
According to the depth information on constituent parts plane, combination obtains the depth information of described pars intermedia image object.
4. method as claimed in claim 3, it is characterized in that, described is base unit with unit plane in the described pars intermedia image object, adopts and calculates the depth information on constituent parts plane in the described pars intermedia image object based on the depth information algorithm of binocular solid geometric relationship, comprising:
Described pars intermedia image object is carried out profile extract operation;
The result who extracts operation according to profile carries out modeling to described pars intermedia image object, obtains the three-dimensional wire-frame model of pars intermedia image object;
According to default unit area each face in the three-dimensional wire-frame model of described pars intermedia image object is divided, obtained the unit plane of each face correspondence in the three-dimensional wire-frame model;
Adopt the depth calculation mode of binocular solid geometric relationship, unit in the three-dimensional wire-frame model is calculated on the plane, obtain the depth information on each unit plane.
5. the method for claim 1 is characterized in that, if comprise the sky image object in the object set of described image, the depth information that then calculates the sky image object in the object set of described image comprises:
The depth information of the sky image object in the object set of described image is labeled as the infinite distance.
6. a depth map generating apparatus is characterized in that, comprising:
Cut apart module, be used for the image of input is carried out cutting operation, obtain the object set of image, comprise in sky image object, ground image object and the pars intermedia image object any one or more in the object set of described image;
Computing module for each the image object selected depth information algorithm the object set that is respectively described image from the depth information algorithm that presets, calculates the depth information of correspondence image object;
Composite module is used for the depth information combination with each image object, obtains the depth map of the image of described input.
7. device as claimed in claim 6 is characterized in that, described computing module comprises:
First selected cell is used for being the depth information algorithm of the ground image Object Selection the object set of described image based on interpolation calculation from the depth information algorithm that presets;
First computing unit is for the depth information that calculates at least three points of ground image object;
First processing unit is used for the depth information according at least three points that calculate, and adopts the mode of interpolation to obtain the depth information of having a few in the described ground image object.
8. device as claimed in claim 6 is characterized in that, described computing module comprises:
Second selected cell is used for being the depth information algorithm of the ground image Object Selection the object set of described image based on the binocular solid geometric relationship from the depth information algorithm that presets;
Second computing unit, being used for described pars intermedia image object unit plane is base unit, employing is calculated the depth information on constituent parts plane in the described pars intermedia image object based on the depth information algorithm of binocular solid geometric relationship, and described unit plane refers to that the area size is the plane of default unit area size in the pars intermedia image object;
Second processing unit, according to the depth information on constituent parts plane, combination obtains the depth information of described pars intermedia image object.
9. device as claimed in claim 8 is characterized in that, described second computing unit comprises:
Extract subelement, described pars intermedia image object is carried out profile extract operation;
The modeling subelement is used for extracting the result who operates according to profile described pars intermedia image object is carried out modeling, obtains the three-dimensional wire-frame model of pars intermedia image object;
Subelement is divided on the plane, is used for according to default unit area each face of the three-dimensional wire-frame model of described pars intermedia image object being divided, and obtains the unit plane of each face correspondence in the three-dimensional wire-frame model;
Computation subunit for the depth calculation mode that adopts the binocular solid geometric relationship, is calculated on the plane unit in the three-dimensional wire-frame model, obtains the depth information on each unit plane.
10. device as claimed in claim 6 is characterized in that, described computing module comprises:
Indexing unit is used for the depth information of the sky image object of the object set of described image is labeled as the infinite distance.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310069640.3A CN103198473B (en) | 2013-03-05 | 2013-03-05 | A kind of degree of depth drawing generating method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310069640.3A CN103198473B (en) | 2013-03-05 | 2013-03-05 | A kind of degree of depth drawing generating method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103198473A true CN103198473A (en) | 2013-07-10 |
| CN103198473B CN103198473B (en) | 2016-02-24 |
Family
ID=48720979
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310069640.3A Active CN103198473B (en) | 2013-03-05 | 2013-03-05 | A kind of degree of depth drawing generating method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103198473B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103617608A (en) * | 2013-10-24 | 2014-03-05 | 四川长虹电器股份有限公司 | Method for obtaining depth map through binocular image |
| CN105374019A (en) * | 2015-09-30 | 2016-03-02 | 华为技术有限公司 | A multi-depth image fusion method and device |
| CN106231177A (en) * | 2016-07-20 | 2016-12-14 | 成都微晶景泰科技有限公司 | Scene depth measuring method, equipment and imaging device |
| CN108024051A (en) * | 2016-11-04 | 2018-05-11 | 宁波舜宇光电信息有限公司 | Distance parameter computational methods, dual camera module and electronic equipment |
| CN108419446A (en) * | 2015-08-24 | 2018-08-17 | 高通股份有限公司 | System and method for the sampling of laser depth map |
| CN109274864A (en) * | 2018-09-05 | 2019-01-25 | 深圳奥比中光科技有限公司 | Depth camera, depth calculation System and method for |
| CN109658451A (en) * | 2018-12-04 | 2019-04-19 | 深圳市道通智能航空技术有限公司 | A kind of depth perception method, device and depth perception equipment |
| CN111025137A (en) * | 2019-12-13 | 2020-04-17 | 苏州华电电气股份有限公司 | Open type isolating switch state sensing device |
| CN111080804A (en) * | 2019-10-23 | 2020-04-28 | 贝壳技术有限公司 | Three-dimensional image generation method and device |
| WO2022040941A1 (en) * | 2020-08-25 | 2022-03-03 | 深圳市大疆创新科技有限公司 | Depth calculation method and device, and mobile platform and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110181589A1 (en) * | 2010-01-28 | 2011-07-28 | The Hong Kong University Of Science And Technology | Image-based procedural remodeling of buildings |
| CN102308320A (en) * | 2009-02-06 | 2012-01-04 | 香港科技大学 | Generating three-dimensional models from images |
-
2013
- 2013-03-05 CN CN201310069640.3A patent/CN103198473B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102308320A (en) * | 2009-02-06 | 2012-01-04 | 香港科技大学 | Generating three-dimensional models from images |
| US20110181589A1 (en) * | 2010-01-28 | 2011-07-28 | The Hong Kong University Of Science And Technology | Image-based procedural remodeling of buildings |
Non-Patent Citations (2)
| Title |
|---|
| DEREK HOIEM: "Putting Objects in Perspective", 《INTERNATIONAL JOURNAL OF COMPUTER VISION》, 17 April 2008 (2008-04-17), pages 3 - 25 * |
| 李乐等: "基于内容理解的单幅静态街景图像深度估计", 《机器人》, vol. 33, no. 1, 31 January 2011 (2011-01-31), pages 174 - 180 * |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103617608B (en) * | 2013-10-24 | 2016-07-06 | 四川长虹电器股份有限公司 | By the method that binocular image obtains depth map |
| CN103617608A (en) * | 2013-10-24 | 2014-03-05 | 四川长虹电器股份有限公司 | Method for obtaining depth map through binocular image |
| CN108419446A (en) * | 2015-08-24 | 2018-08-17 | 高通股份有限公司 | System and method for the sampling of laser depth map |
| US11915502B2 (en) | 2015-08-24 | 2024-02-27 | Qualcomm Incorporated | Systems and methods for depth map sampling |
| CN114637023A (en) * | 2015-08-24 | 2022-06-17 | 高通股份有限公司 | System and method for laser depth map sampling |
| CN108419446B (en) * | 2015-08-24 | 2022-03-25 | 高通股份有限公司 | System and method for laser depth map sampling |
| CN105374019A (en) * | 2015-09-30 | 2016-03-02 | 华为技术有限公司 | A multi-depth image fusion method and device |
| WO2017054589A1 (en) * | 2015-09-30 | 2017-04-06 | 华为技术有限公司 | Multi-depth image fusion method and apparatus |
| CN105374019B (en) * | 2015-09-30 | 2018-06-19 | 华为技术有限公司 | A kind of more depth map fusion methods and device |
| CN106231177A (en) * | 2016-07-20 | 2016-12-14 | 成都微晶景泰科技有限公司 | Scene depth measuring method, equipment and imaging device |
| CN108024051B (en) * | 2016-11-04 | 2021-05-04 | 宁波舜宇光电信息有限公司 | Distance parameter calculation method, double-camera module and electronic equipment |
| CN108024051A (en) * | 2016-11-04 | 2018-05-11 | 宁波舜宇光电信息有限公司 | Distance parameter computational methods, dual camera module and electronic equipment |
| CN109274864A (en) * | 2018-09-05 | 2019-01-25 | 深圳奥比中光科技有限公司 | Depth camera, depth calculation System and method for |
| CN109658451A (en) * | 2018-12-04 | 2019-04-19 | 深圳市道通智能航空技术有限公司 | A kind of depth perception method, device and depth perception equipment |
| CN111080804B (en) * | 2019-10-23 | 2020-11-06 | 贝壳找房(北京)科技有限公司 | Three-dimensional image generation method and device |
| CN111080804A (en) * | 2019-10-23 | 2020-04-28 | 贝壳技术有限公司 | Three-dimensional image generation method and device |
| CN111025137A (en) * | 2019-12-13 | 2020-04-17 | 苏州华电电气股份有限公司 | Open type isolating switch state sensing device |
| WO2022040941A1 (en) * | 2020-08-25 | 2022-03-03 | 深圳市大疆创新科技有限公司 | Depth calculation method and device, and mobile platform and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103198473B (en) | 2016-02-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103198473A (en) | Depth image generating method and device | |
| WO2019127445A1 (en) | Three-dimensional mapping method, apparatus and system, cloud platform, electronic device, and computer program product | |
| Wendel et al. | Dense reconstruction on-the-fly | |
| Liang et al. | Image based localization in indoor environments | |
| CN107392958B (en) | Method and device for determining object volume based on binocular stereo camera | |
| CN111141264B (en) | Unmanned aerial vehicle-based urban three-dimensional mapping method and system | |
| EP2385496A1 (en) | Extraction of 2D surfaces from a 3D point cloud | |
| CN109186551B (en) | Method and device for extracting characteristic points of oblique photogrammetry building and storage medium | |
| KR20140060454A (en) | Backfilling points in a point cloud | |
| CN109269472B (en) | Method and device for extracting characteristic line of oblique photogrammetry building and storage medium | |
| EP3274964B1 (en) | Automatic connection of images using visual features | |
| CN107851329B (en) | Displaying objects based on multiple models | |
| CN112789609A (en) | Map updating method and device, movable platform and storage medium | |
| CN110132242B (en) | Triangularization method for multi-camera instant positioning and map construction and moving body thereof | |
| JP2020518918A (en) | Information processing method, apparatus, cloud processing device, and computer program product | |
| CN111721281B (en) | Position identification method and device and electronic equipment | |
| CN105466399B (en) | Quickly half global dense Stereo Matching method and apparatus | |
| WO2023016082A1 (en) | Three-dimensional reconstruction method and apparatus, and electronic device and storage medium | |
| CN109255809A (en) | A kind of light field image depth estimation method and device | |
| KR20130137076A (en) | Device and method for providing 3d map representing positon of interest in real time | |
| CN105444773A (en) | Navigation method and system based on real scene recognition and augmented reality | |
| Mejia et al. | Vehicle speed estimation using computer vision and evolutionary camera calibration | |
| Feng et al. | Object detection and localization based on binocular vision for autonomous vehicles | |
| Rothermel et al. | Fast and robust generation of semantic urban terrain models from UAV video streams | |
| CN116958485A (en) | Visual field analysis method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20180117 Address after: The South Road in Guangdong province Shenzhen city Fiyta building 518000 floor 5-10 Nanshan District high tech Zone Patentee after: Shenzhen Tencent Computer System Co., Ltd. Address before: Shenzhen Futian District City, Guangdong province 518057 Zhenxing Road, SEG Science Park 2 East Room 403 Patentee before: Tencent Technology (Shenzhen) Co., Ltd. |
|
| TR01 | Transfer of patent right |