JP4999589B2 - Image processing apparatus and method - Google Patents

Description

The present invention relates to an image processing apparatus and method.

  In recent years, video recording / reproducing apparatuses for recording and reproducing digital television broadcast programs such as BS digital broadcast or terrestrial digital broadcast on a disk medium such as a hard disk or an optical disk have become widespread. In addition, video recording / playback apparatuses that are compatible with networks and that can transmit and receive moving image data to and from personal computers and digital home appliances on a LAN (Local Area Network) have been commercialized.

  DLNA (Digital Living Network Alliance) is a mechanism for sharing moving image data on a home network, and products that comply with the implementation guidelines established by DLNA are increasing.

In recent years, with the increase in capacity of hard disks, a large amount of television broadcast programs and moving image data can be recorded. For example, when the hard disk has a capacity of 250 GB (gigabytes), it is possible to record a television broadcast of 100 hours or more in the standard image quality mode and 200 hours or more in the long-time mode. As a result, the number of programs that can be stored in the apparatus is enormous, from tens to hundreds. Patent Document 1 describes a technique for easily searching for a desired program from such a large number of recorded programs.
JP 2002-359803 A

  Conventional search techniques search for a desired program based on character information, and are still bothersome. In addition, it cannot be applied to moving image data captured by a user with a digital camera or a digital video camera unless the user inputs character information.

  Therefore, there is still a problem that it takes a lot of time and effort to search for content that the user wants to watch from a large amount of moving image data recorded on the hard disk.

An object of the present invention is to provide an image processing apparatus and method capable of quickly retrieving a desired moving image from a large number of moving images.

The image processing apparatus according to the present invention detects recording periods for recording a moving image on a recording medium, and appearance periods of a plurality of specific objects included in the moving image, and the appearance periods for each of the plurality of specific objects Generating means for generating object related information, acquiring means for acquiring an image, detecting means for detecting a specific object from the image acquired by the acquiring means, the object related information and a detection result of the detecting means Based on the moving image recorded on the recording medium, a playlist generating unit that generates a playlist for selecting and reproducing the appearance period of the specific object detected by the detecting unit, and the detecting unit Is the first of the plurality of specific objects included in the moving image from the one-screen image acquired by the acquiring means. When the specific object and the second specific object are detected, the playlist generation means selects and reproduces the period in which the first specific object and the second specific object appear together A first play list for selecting, a second play list for selecting and reproducing the appearance period of the first specific object, and selecting and reproducing the appearance period of the second specific object Generating a third play list to be performed .

An image processing method according to the present invention is a method for processing a moving image recorded on a recording medium, wherein the appearance period of a plurality of specific objects included in the moving image is detected, and each of the plurality of specific objects is detected. Generating object-related information relating to the appearance period, an acquisition step of acquiring an image, a detection step of detecting a specific object from the acquired image, the object-related information, and a detection result of the detection step And generating a playlist for selecting and reproducing the appearance period of the specific object detected by the detection step from the moving images recorded on the recording medium, the detection step comprising: From the acquired one-screen image, a first of a plurality of the specific objects included in the moving image When the predetermined object and the second specific object are detected, the step of generating the playlist selects a period in which the first specific object and the second specific object appear together. A first playlist for reproduction, a second playlist for selecting and reproducing an appearance period of the first specific object, and an appearance period of the second specific object A third play list for reproduction is generated .

  According to the present invention, even when a large amount of moving images are stored in a recording medium, it is possible to quickly provide a user with a moving image showing a specific object, for example, a specific person, as a reproduction candidate. As a result, it is possible for the user to quickly reproduce content that the user is likely to view, and the time and labor for searching for a desired moving image from a large amount of moving images can be reduced.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

  FIG. 1 shows a schematic block diagram of an embodiment of a recording / reproducing apparatus according to the present invention, and FIG. 2 shows an example of connection with peripheral devices.

  The recording / reproducing apparatus 10 which is one embodiment of the present invention is connected to a camera 12 as an imaging means via a USB cable or an IEEE1394 cable. In this embodiment, the camera 12 is a video camera that outputs a moving image, but may be a digital still camera that outputs a one-shot shot image.

  A video camera 16 is further connected to the recording / reproducing apparatus 10 via the LAN 14. In this embodiment, moving images and audio output from the video camera 16 are supplied to the recording / reproducing apparatus 10 via the LAN 14 and recorded on the recording medium of the recording / reproducing apparatus 10. Unless specifically excluded, the moving image includes sound. The video / audio output of the recording / reproducing apparatus 10 is supplied to the video / audio monitor 18. The video / audio monitor 18 displays the playback video from the recording / playback device 10 on the screen of the video display device 18a, and outputs the playback audio from the speaker 18b.

  The configuration and basic operation of the recording / reproducing apparatus 10 will be described. A moving image from the video camera 16 is input to the communication processing device 20 of the recording / reproducing device 10 via the LAN 14. The communication processing device 20 can receive data from the LAN 14 by a known protocol such as HTTP (Hyper Text Transfer Protocol) or FTP (File Transfer Protocol). Assume that the video camera 16 and the communication processing device 20 are set with appropriate IP addresses and know the IP addresses of each other. For example, a well-known UPnP (Universal Play and Play) standard function is used.

  The recording / reproducing apparatus 10 includes a user interface (UI) 22 including operation keys, operation buttons, a remote control device, and the like operated by a user. The user can use the user interface 22 to instruct the recording / reproducing apparatus 10 about the operation mode and the start / end of recording / reproduction.

  In the recording mode, the communication processing device 20 supplies the moving image from the video camera 16 to the recording processing device 24 and the face recognition processing device 26. When the video camera 16 compresses and encodes a moving image and transmits it to the LAN 14 for moving image transmission via the LAN 14, it is clear that the recording / reproducing apparatus 10 has a corresponding image expansion device (not shown). It is.

  In accordance with the recording start instruction, the recording processing device 24 records the moving image from the communication processing device 20 on a hard disk (HDD) 28 that is a recording medium. When a specific image compression method is used for moving image recording on the HDD 28, the recording processing device 24 includes an image compression device for that purpose.

  The face recognition processing device 26 recognizes the face of a person by image recognition from the frame image of the moving image from the communication processing device 20 or the frame image of a fixed frame period, and extracts a face feature amount (first object feature amount). . This corresponds to the extraction of the first object feature amount by the first object recognition process. For example, the face recognition processing device 26 recognizes a face at a cycle of 10 frames or a cycle of 1/5 of the frame rate. The face recognition processing device 26 performs face recognition when the face size is equal to or larger than a predetermined size. Various known methods can be applied to the face recognition processing technology of the face recognition processing device 26. For example, a method may be used in which the contour of the face is detected by edge detection, and the positions of eyes, nose and mouth are extracted as feature amounts. The face recognition processing device 26 supplies the extracted face feature amount to the face discrimination processing device 30.

  The face discrimination processing device 30 searches the face feature amount database 32 using the face feature amount from the face recognition processing device 26 as a key, and discriminates a person. The face feature amount database 32 includes a record that associates a face feature amount with a person (actually, a face identifier that identifies a person). The face discrimination processing device 30 stores information indicating a person who has been discriminated as a result of the search of the face feature amount database 32, that is, a person identifier in the management table 34. The face feature amount database 32 and the management table 34 may be stored in the HDD 28 or may be stored in a storage medium different from the HDD 28.

  In the management table 34, information indicating which person is included in which period in the moving image finally supplied from the video camera 16 and recorded in the HDD 28 is stored in the appearance unit of the person. That is, each row record indicates a period during which a person identified by the face identifier in the face identifier field 42 is shown in the moving image specified in the content name field. In this regard, the management table 34 is an object history table that stores the appearance history of persons (objects) included in a moving image.

  FIG. 3 shows an example of the structure of the management table 34. The management table 34 includes a discrimination identifier field 40, a face identifier field 42, a content name field 44, a start time field 46, an end time field 48, and a time field 50. The management table 34 describes in which period of which content a specific object (a person's face in this embodiment) appears in a moving image recorded in the HDD 28, and the object related information in the claims Corresponding to The function of the face discrimination processing device 30 that generates the management table 34 corresponds to the object-related information generation unit in the claims.

  The discrimination identifier in the discrimination identifier field 40 is a unique identifier generated by the face discrimination processing device 30 when a new face is recognized in the face recognition process or when a face is recognized again. The face identifier in the face identifier field 42 is an identifier that uniquely identifies a face. When the face feature amount from the face recognition processing device 26 is already registered in the face feature amount database 32, the face discrimination processing device 30 reads a face identifier corresponding to the face feature amount from the face feature amount database 32, and the face identifier Store in field 42.

  The content name in the content name field 44 is a name given to the moving image data received from the video camera 16 by an application (not shown). That is, the face discrimination processing device 30 stores a content name such as a file name provided from an application (not shown) in the content name field 44 of the management table 34.

  The start time field 46 stores the time code when the face appears for the first time or again in the frame image, and the end time field 48 stores the time code when the face is no longer recognized. The time code is an elapsed time when the first frame image of the moving image data is set to 00:00:00, and the value increases in accordance with the frame rate of the moving image data. For example, when the frame rate is 30 frames / second and the face recognition process is performed every 6 frames, the time code increases by 1 second every 5 frames. That is, the person specified by the face identifier in the face identifier field 42 from the start time in the start time field 46 to immediately before the end time in the end time field 48 is included in the moving image being recorded. Become.

  The time field 50 stores the period during which the person specified in the face identifier field 42 has appeared in the moving image, specifically, the result of subtracting the value of the start time field 46 from the value of the end time field 48. The

  Although the operation of recording a moving image from the video camera 16 on the HDD 28 has been described, a one-shot moving image from the video camera 16, that is, a still image can be recorded on the HDD 28 in the same process. Of course, many recent video cameras can shoot still images with a resolution that is high enough to be used as a digital still camera.

  In addition, the operation of executing the face recognition process and the face discrimination process simultaneously with the recording of the moving image has been described, but the face recognition process and the face discrimination process may be executed after the recording of the moving image, for example, in a standby state. Good.

  The recording / reproducing apparatus 10 of the present embodiment can reproduce any image (moving image or still image) recorded in the HDD 28 in the reproduction mode. In the present embodiment, in particular, a playlist of moving images in which a specific person is photographed can be created, and recorded images on the HDD 28 can be reproduced according to the playlist. That is, it is possible to select and play a moving image of a portion specified in the play list from the moving images recorded in the HDD 28. This operation mode is called a playlist creation mode.

  In the playlist creation mode, the output image signal of the camera 12 is supplied to the face recognition processing device 26 via the digital interface 38. The digital interface 38 is, for example, an interface conforming to USDB or IEEE1394, or an image capture device that converts an analog image signal into a digital image signal. The face recognition processing device 26 recognizes the face of the person being shown from the captured image of the camera 12 as in the recording mode. Of course, when the camera 12 is a still camera, the face recognition processing device 26 recognizes a face from a one-shot photographed image (still image) from the camera 12. The face recognition processing device 26 supplies the face feature amount (second object feature amount) extracted by face recognition to the face discrimination processing device 30. This corresponds to the extraction of the second object feature amount by the second object recognition process.

  The face discrimination processing device 30 searches the face feature value database 32 using the face feature value from the face recognition processing device 26 as a search key, and further uses the face identifier of the person matched in the search of the face feature value database 32 as a search key. The management table 34 is searched. The face discrimination processing device 30 finally generates a playlist indicating content including a person photographed by the camera 12 from the management table 34 and supplies the playlist to the reproduction processing device 36. That is, the face discrimination processing device 30 corresponds to a playlist generation unit in the claims.

  The reproduction processing device 36 refers to the playlist from the face discrimination processing device 30 to notify the user of the existence of the playlist, and reproduces the corresponding content from the HDD 28 automatically or according to the user's instruction. The reproduction processing device 36 supplies the reproduction video signal and the reproduction audio signal to the video / audio monitor 18.

  FIG. 4 shows an operation flowchart of moving image recording and face recognition according to this embodiment. An application (not shown) is activated, and transmission of moving image data from the video camera 16 to the recording / reproducing apparatus 10 is started using, for example, an HTTP POST request. The communication processing device 20 receives the moving image data from the video camera 16 (S1). At this time, the recording / reproducing apparatus 10 assigns a content name to the received moving image data. In the present embodiment, for example, MOVIE-X. MPG, and if it is still image data, IMAGE-X. JPG. X is a serial number.

  As described above, the face recognition processing device 26 extracts frame images at predetermined intervals from the moving image data from the communication processing device 20, and performs face recognition processing (S2). When the face is recognized in the frame image (S3), the extracted face feature amount is supplied to the face discrimination processing device 30, and when the face is not recognized in the frame image (S3), the face recognition process is ended.

  The face discrimination processing device 30 indicates whether or not the facial feature amount from the facial recognition processing device 26 is registered in the facial feature amount database 32 (S4). If the extracted face feature value is not registered in the face feature value database 32 (S4), a new unique face identifier is generated and registered in the face feature value database 32 together with the face feature value (S5). On the other hand, when the extracted face feature value is registered in the face feature value database 32 (S4), a face identifier corresponding to the face feature value is acquired from the face feature value database 32 (S6).

  The generated face identifier (S5) or the acquired face identifier (S6) is recorded in the management table 34 (S7). When a face is newly recognized, a new record in which a discrimination identifier field 40, a face identifier field 42, a content name field 44, and a start time field 46 are written is added to the management table 34. A time code when the face is no longer recognized is written in the end time field 48 of the new record, and an elapsed time from the start time to the end time is written in the time field 50. Each record in the management table 34 indicates a period during which the person identified by the face identifier in the face identifier field 42 is shown in the moving image.

  The above processing (S2 to S7) is repeated until the reception of the moving image data is completed (S8). When the reception of moving image data is completed, the flow shown in FIG. 4 ends.

  5, 6 and 7 show examples of the contents and transition of the management table 34. FIG. FIG. 5 shows an example of the contents of the management table 34 when the recording / reproducing apparatus 10 receives and records moving image data (content name: MOVIE-1.MPG). FIG. 6 shows an example of the contents of the management table 34 when still image data (content name: IMAGE-1.JPG) is received from the video camera 16 and recorded, following FIG. FIG. 7 shows an example of the contents of the management table 34 when moving image data (content name: MOVIE-2.MPG) is received from the video camera 16 and recorded, following FIG. The face discrimination processing device 30 updates the management table 34 as shown in FIGS.

  The operation in the playlist reproduction mode for creating a playlist of images in which the person is photographed using the person photographed by the camera 12 as a search key will be described. FIG. 8 shows an operation flowchart of the playlist creation mode.

  An application (not shown) takes captured image data of the camera 12 (S11). Thereby, the captured image data of the camera 12 is supplied to the face recognition processing device 26 via the digital interface 38. When the IEEE 1394 interface is used for the digital interface 38, the IEEE 1394 AV / C protocol can be used. As in the recording mode, the face recognition processing device 26 recognizes a person's face from the captured image data of the camera 12 (S12).

  When the face is recognized (S13), the face recognition processing device 26 supplies the extracted face feature amount to the face discrimination processing device 30. When the face cannot be recognized (S13), the face recognition processing device 26 ends the face recognition processing and proceeds to S17.

  The face discrimination processing device 30 searches the face feature amount database 32 using the face feature amount from the face recognition processing device 26 as a search key (S14). When the face feature amount extracted from the captured image data of the camera 12 has not been registered in the face feature amount database 32 (S14), the face discrimination processing device 30 ends the face discrimination processing and proceeds to S17.

  When the face feature amount extracted from the captured image data of the camera 12 has already been registered in the face feature amount database 32 (S14), the face discrimination processing device 30 sets the face identifier corresponding to the face feature amount to the face feature amount database 32. Read from. Then, the face discrimination processing device 30 searches the management table 34 using the face identifier from the face feature amount database 32 as a search key, and generates a playlist of images including this face identifier from the management table 34 (S15).

  For example, it is assumed that the management table 34 has the contents shown in FIG. 7, and “F-01” is acquired as the face identifier in S14. In this case, the face discrimination processing device 30 filters the management table 34 shown in FIG. 7 with face identifier = “F-01”, and generates the playlist shown in FIG. That is, a record with face identifier = “F-01” is extracted from the management table 34.

  The face discrimination processing device 30 notifies the playback processing device 36 of the playlist generation event together with the playlist. In response to this event, the playback processing device 36 notifies the user of the presence of the playlist (S16). Specifically, the playback processing device 36 displays the presence of the playlist on the screen of the video display device 18a, or outputs the presence of the playlist from the speaker 18b by voice. FIG. 10 shows a display example of the video display device 18a. In the example shown in FIG. 10, when a “play” button for playing a playlist is displayed on the video display device 18a, and the user instructs playback using the user interface 22, the playback processing device 36 In response to the instruction, playback of the playlist is started.

  Steps S12 to S16 are repeated until the capturing of the video signal from the camera 12 is completed due to an instruction from the user or a timeout (S17).

  When there is no instruction from the user for a certain period of time, images described in the playlist may be automatically reproduced in order according to the playlist.

  In this way, in this embodiment, a playlist including a person photographed by the camera 12 can be automatically generated from the recorded image, and the playlist can be executed in response to a user instruction or automatically.

  Although an embodiment has been described in which a moving image sent via a LAN is a target for recording and face recognition, it is obvious that a video signal broadcast on television can be a target for recording and face recognition. Further, the recording / reproducing apparatus 10 may be integrated with the video / audio monitor 18.

  The face of a person is exemplified as a key for creating a playlist. However, it is obvious that the present invention can be applied to other objects in general, for example, specific animals such as dogs and cats, specific landscapes, and the like. . In this regard, the face recognition processing device 26 is an example of an object recognition processing unit.

  In the flow shown in FIG. 8, a playlist of one person photographed by the camera 12 is created. By using a device that operates at high speed as the face recognition processing device 26, it is possible to recognize the faces of a plurality of people within one screen. In that case, not only a playlist of images showing each of a plurality of people but also a playlist of images showing the plurality of people at the same time is created. FIG. 11 shows a flowchart of the operation.

  An application (not shown) captures captured image data of the camera 12 (S21). Thereby, the captured image data of the camera 12 is supplied to the face recognition processing device 26 via the digital interface 38. As in the recording mode, the face recognition processing device 26 recognizes a human face from the captured image data of the camera 12 (S22). When multiple people are shown, the faces of multiple people are recognized.

  For each recognized face, the face recognition processing device 26 supplies the extracted face feature amount to the face discrimination processing device 30 (S23), and the face discrimination processing device 30 checks whether it has been registered in the face feature amount database 32. (S24). When the face feature value has been registered in the face feature value database 32 (S24), the face discrimination processing device 30 reads the face identifier corresponding to each registered face feature value from the face feature value database 32.

  The face discrimination processing device 30 searches the management table 34 using each face identifier read from the face feature quantity database 32 as a search key, a play list of images including these face identifiers alone, and a play including all face identifiers. A list is created (S25).

  For example, it is assumed that the management table 34 has the contents shown in FIG. 7 and that “F-01” and “F-02” are acquired as face identifiers in S24. In this case, the face discrimination processing device 30 creates the playlist shown in FIGS. 12 and 13 in addition to the playlist shown in FIG. 9 for the face identifier = “F-01”. FIG. 12 shows a play list for face identifier = “F-02”. FIG. 13 shows a playlist that simultaneously includes both “F-01” and “F-02” as face identifiers.

  In the playlist shown in FIG. 13, the start time, end time, and time are defined as a period in which the face identifier = “F-01” is reflected and a period in which the face identifier = “F-02” is reflected and the periods overlap. To do. That is, the playlist shown in FIG. 13 corresponds to the logical product of the playlist shown in FIG. 10 and the playlist shown in FIG.

  The face discrimination processing device 30 notifies the playback processing device 36 of the playlist generation event together with the playlist. In response to this event, the playback processing device 36 notifies the user of the presence of the playlist (S26). Specifically, the playback processing device 36 displays the presence of the playlist on the screen of the video display device 18a, or outputs the presence of the playlist from the speaker 18b by voice. FIG. 14 shows a display example of the video display device 18a. A message 60 for the playlist shown in FIG. 10, a message 62 for the playlist shown in FIG. 12, and a message 64 for the playlist shown in FIG. 13 are simultaneously displayed. Other configurations are the same as those in FIG.

  Steps S22 to S26 are repeated until the capturing of the video signal from the camera 12 is completed due to an instruction from the user or a timeout (S27).

  When there is no instruction from the user for a certain period of time, images described in the playlist may be automatically reproduced in order according to the playlist.

  In this way, in this embodiment, a playlist including a plurality of persons photographed simultaneously by the camera 12 can be automatically generated from the recorded image. It is clear that a playlist that always includes one of a plurality of persons can be generated by the logical sum of the playlists for each person.

It is a schematic block diagram of one Example of this invention. It is a connection block diagram with the peripheral device of a present Example. 4 is a diagram showing a configuration of a management table 34. FIG. It is an operation | movement flowchart at the time of the moving image recording of a present Example. It is an example of the contents of a management table. 6 is a content example of a management table following FIG. FIG. 7 is a content example of a management table following FIG. 6. FIG. It is an operation | movement flowchart in play list creation mode. It is an example of a created playlist. It is an example of a display of the play list creation message of step S16. It is an operation | movement flowchart in the play list creation mode corresponding to the several person image | photographed simultaneously. It is an example of a play list generated by the flow shown in FIG. FIG. 12 is an example of an image playlist including a plurality of people generated in the flow shown in FIG. It is an example of a display of the play list creation message of step S26.

Explanation of symbols

10: Recording / reproducing apparatus 12: Camera 14: LAN
16: Video camera 18: Video / audio monitor 18a: Video display device 18b: Speaker 20: Communication processing device 22: User interface (UI)
24: Recording processing device 26: Face recognition processing device 28: Hard disk (HDD)
30: Face discrimination processing device 32: Face feature quantity database 34: Management table 36: Playback processing device 38: Digital interface 40: Discrimination identifier field 42: Face identifier field 44: Content name field 46: Start time field 48: End time field 50: Time field

Claims (6)

Recording means for recording a moving image on a recording medium;
Means for detecting an appearance period of a plurality of specific objects included in the moving image, and generating object related information regarding the appearance period for each of the plurality of specific objects;
An acquisition means for acquiring an image;
Detecting means for detecting a specific object from the image acquired by the acquiring means;
Based on the object-related information and the detection result of the detection means, a playlist for selecting and reproducing the appearance period of the specific object detected by the detection means from the moving images recorded on the recording medium. Playlist generating means for generating ,
The detection unit detects a first specific object and a second specific object among a plurality of the specific objects included in the moving image from an image of one screen acquired by the acquisition unit. In this case, the playlist generating means selects a first playlist for reproduction by selecting a period in which the first specific object and the second specific object appear together, and the first specific object. A second playlist for selecting and reproducing the appearance period of the second object and a third playlist for selecting and reproducing the appearance period of the second specific object are generated.
An image processing apparatus.   The object related information includes information on an appearance start time and an appearance stop time for each of the plurality of objects, and the playlist generation unit is configured to start appearance of a specific object detected by the detection unit included in the object related information. The image processing apparatus according to claim 1, wherein the playlist is generated based on time and appearance stop time information.   The detecting means detects the specific object included in the moving image recorded on the recording medium when the moving image is recorded, and the means for generating the object related information is a moving image recorded on the recording medium. The image processing apparatus according to claim 1, wherein the object related information is generated based on a detection result of the specific object included in an image. Comprises means for generating a feature amount information including the identifier of the previous SL plurality of said specific object each feature quantity included in the moving image recorded on a recording medium and a plurality of said specific object,
The means for generating the object related information generates the object related information in which a plurality of identifiers of the specific object and information indicating the appearance period for each of the specific objects are associated with each other.
The playlist generation unit detects an identifier of the specific object detected by the detection unit based on the feature amount of the specific object detected by the detection unit and the feature amount information, and corresponds to the detected identifier. The image processing apparatus according to claim 1, wherein the play list is generated by detecting an appearance period to be detected from the object-related information. Face recognition means for recognizing a person's face as the specific object, and the means for generating the object related information, the appearance of the specific object based on a face recognition result from the moving image by the face recognition means A period of time is detected, and the detection unit includes the identification included in the one-screen image acquired by the acquisition unit based on a face recognition result from the one-screen image acquired by the acquisition unit by the face recognition unit. 5. The image processing apparatus according to claim 1 , wherein the object is detected . A method of processing a moving image recorded on a recording medium,
Detecting an appearance period of a plurality of specific objects included in the moving image, and generating object-related information regarding the appearance period for each of the plurality of specific objects;
An acquisition step of acquiring an image;
A detecting step of detecting a specific object from the acquired image;
Based on the object related information and the detection result of the detection step, a playlist for selecting and reproducing the appearance period of the specific object detected by the detection step from the moving images recorded on the recording medium. Step to generate
And
When the detection step detects a first specific object and a second specific object among the plurality of specific objects included in the moving image from the acquired one-screen image, The step of generating a playlist includes: a first playlist for selecting and reproducing a period in which the first specific object and the second specific object appear together; and the first specific object A second playlist for selecting and reproducing the appearance period of the object and a third playlist for selecting and reproducing the appearance period of the second specific object are generated.
An image processing method.

Images