Image processing device, video search system, image processing method, and program

Abstract

The present invention makes it possible to search for a video, in which a desired object is captured with high image quality, while reducing the amount of data. This image processing device includes: an object detection unit that detects, from a video, an object included in a label list which includes one or more objects; a video compression unit that compresses the video so that the image quality of a region of the detected object in the video becomes higher than the image quality of a region other than the region of the detected object in the video; and a list update unit that updates the label list on the basis of the history of search keywords designated by a user in a video search for searching for the video by using the search keywords.

Description

Image processing device, video search system, image processing method, and program

[0001] The present disclosure relates to an image processing device, a video search system, an image processing method, and a program.

[0002] As a related technique, Patent Document 1 discloses a video processing device. The video processing device described in Patent Document 1 detects an object from a camera image. Detecting an object includes recognizing the type of the object. Based on the object detection result, the video processing device selects, for example, an area of ​​an object having a predetermined object type as a gaze area. The video processing device increases the image quality of the selected gaze area and decreases the image quality of other areas in the camera image. The video processing device transmits the camera image to a remote monitoring device. The video processing device described in Patent Document 1 can reduce the bit rate of the camera image transmitted to the remote monitoring device by maintaining the image quality of the gaze area including a person or object while decreasing the image quality of the surrounding area.

[0003] International Publication No. 2024 / 047794

[0004] Consider a case where a video search is performed on videos whose bit rate has been reduced using the video processing device described in Patent Document 1. A user provides a search system with a search keyword containing a desired object in order to find a video showing the desired object. However, if the desired object is not set in an area where high image quality is to be achieved, the desired object will be included in an area of ​​low image quality, and the probability of finding a video showing the desired object in high image quality will be low.

[0005] One of the objectives of the present disclosure is to provide a video search system, an image processing device, an image processing method, and a program that enable searching for videos that show desired objects in high image quality while reducing the amount of data.

[0006] An image processing device according to a first aspect of the present disclosure includes an object detection unit that detects from a video an object included in a label list that includes one or more objects; a video compression unit that compresses the video so that the image quality of an area of ​​the video where the detected object is located is higher than the image quality of areas of the video other than the area of ​​the detected object; and a list update unit that updates the label list based on a history of search keywords specified by a user in a video search that searches the video using search keywords.

[0007] A video search system according to a second aspect of the present disclosure includes the image processing device described above and a video search device that searches for video related to a search keyword specified by a user from a video storage unit that stores the compressed video, based on the search keyword.

[0008] An image processing method according to a third aspect of the present disclosure includes detecting, from a video, an object included in a label list including one or more objects, compressing the video so that an image quality of an area of ​​the video corresponding to the detected object is higher than an image quality of an area of ​​the video other than the area of ​​the detected object, and updating the label list based on a history of search keywords specified by a user in a video search in which the video is searched using the search keywords.

[0009] A program according to a fourth aspect of the present disclosure causes a computer to execute a process of detecting, from a video, an object included in a label list including one or more objects, compressing the video so that the image quality of an area in the video where the detected object is located is higher than the image quality of an area in the video other than the area where the detected object is located, and updating the label list based on a history of search keywords specified by a user in a video search in which the video is searched using search keywords.

[0010] The video search system, image processing device, image processing method, and program according to the present disclosure can reduce the amount of data while enabling a search for videos showing a desired object in high image quality.

[0011] Fig. 1 is a block diagram showing a schematic configuration example of a video search system according to the present disclosure; Fig. 2 is a block diagram showing a configuration example of a video search system according to the present disclosure; Fig. 3 is a block diagram showing a configuration example of a video search device; Fig. 4 is a block diagram showing a configuration example of an image processing device; Fig. 5 is a flowchart showing an operation procedure of an image processing device; Fig. 6 is a block diagram showing a configuration example of a computer device.

[0012] 1 is a block diagram showing a schematic configuration example of a video search system according to the present disclosure. The video search system 10 includes an image processing device 11 and a video search device 30.

[0013] The image processing device 11 has an object detection unit 12, a video compression unit 13, and a list update unit 14. The object detection unit 12 detects from a video an object included in a label list including one or more objects. The video compression unit 13 compresses the video so that the image quality of a region of the video where the detected object is located is higher than the image quality of a region of the video other than the region of the detected object.

[0014] The video storage unit 20 stores the video compressed by the video compression unit 13. The video search device 30 searches the video storage unit 20 for videos related to a search keyword designated by a user.

[0015] The list update unit 14 acquires a history of search keywords designated by a user in a video search in the video search device 30. The list update unit 14 updates the label list used in the video compression unit 13 based on the acquired history of search keywords.

[0016] In the present disclosure, the video compression unit 13 compresses the video so that the area of ​​an object detected by the object detection unit 12 and included in the label list has higher image quality than other areas. The list update unit 14 updates the label list based on the history of search keywords in the video search device 30. For example, the list update unit 14 can add objects that are frequently included in the search keywords to the label list. In this way, it becomes possible to search for videos that show a desired object in high image quality during video search while reducing the amount of video data stored in the video storage unit 20.

[0017] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. Note that the following description and drawings have been omitted and simplified as appropriate for clarity of explanation. In addition, in the following drawings, the same or similar elements are designated by the same reference numerals, and duplicate explanations are omitted as necessary.

[0018] Fig. 2 is a block diagram showing an example configuration of a video search system according to the present disclosure. One embodiment will be described with reference to Fig. 2. The video search system 100 shown in Fig. 2 includes an image processing device 110, a video storage unit 130, a video search device 140, and one or more cameras 210. Below, an example will be described in which a user searches for desired driving video from driving videos collected from multiple vehicles 200 in the video search system 100. The video search system 100 corresponds to the video search system 10 shown in Fig. 1.

[0019] Each of the one or more cameras 210 is mounted on, for example, a vehicle 200. The vehicle 200 is configured as a land vehicle that travels on a road, such as a car, a bus, a taxi, or a truck. The camera 210 captures an image of, for example, the area in front of the vehicle. The number of cameras 210 mounted on each vehicle 200 is not limited to one. Each vehicle 200 may be equipped with multiple cameras 210.

[0020] The image processing device 110 is connected to one or more vehicles 200 via a network 150. The network 150 includes, for example, a wireless communication network using a communication line standard such as Long Term Evolution (LTE). The network 150 may also include a wireless communication network such as Wi-Fi (registered trademark) or a fifth-generation mobile communication system. The image processing device 110 collects images captured by cameras 210 from one or more vehicles 200 via the network 150. The image processing device 110 corresponds to the image processing device 11 shown in FIG. 1 .

[0021] The image processing device 110 compresses the collected video and stores it in the video storage unit 130. The video storage unit 130 is configured as a storage device such as a solid-state drive (SSD) or a hard disk drive (HDD). The video storage unit 130 may be cloud storage. The video storage unit 130 corresponds to the video storage unit 20 shown in FIG. 1.

[0022] The video search device 140 searches the video storage unit 130 for videos related to a search keyword specified by a user. For example, a user such as a developer of autonomous driving artificial intelligence (AI) inputs a search keyword including a desired object into the video search device 140. The video search device 140 searches for driving videos that show the desired object according to the search keyword. The video search device 140 corresponds to the video search device 30 shown in FIG. 1 .

[0023] 3 is a block diagram showing an example configuration of the video search device 140. The video search device 140 has a search keyword input unit 141 and a search unit 142. The video search device 140 is configured as a computer device having, for example, one or more memories and one or more processors. At least some of the functions of each unit in the video search device 140 can be realized by the processor operating in accordance with a program read from the memory.

[0024] The search keyword input unit 141 inputs search keywords specified by the user to the search unit 142. The search keywords include the names of objects included in the video to be searched. The user may input a search query that describes the scene to be searched for in natural language to the video search device 30. The user may input text such as "car collision with truck" or "bicycle running into the road" as the search query to the video search device 140. In this case, the search keyword input unit 141 may extract search keywords from the search query and input the extracted search keywords to the search unit 142.

[0025] The search unit 142 searches the video storage unit 130 and retrieves one or more videos that include the object specified by the user in the search keyword. The search unit 142 displays the search results on a display screen of a display device (not shown). The user selects a displayed video from the search results. The video search device 140 displays the video selected by the user on the display screen.

[0026] 4 is a block diagram showing an example configuration of the image processing device 110. The image processing device 110 includes a video acquisition unit 111, an object detection unit 112, a video compression unit 113, a label list 114, a label assignment unit 115, a search history acquisition unit 116, and a list update unit 117. The image processing device 110 is configured as a computer device including, for example, one or more memories and one or more processors. At least some of the functions of each unit in the image processing device 110 can be realized by the processor operating in accordance with a program read from the memory.

[0027] The video acquisition unit 111 acquires video captured by the camera 210 via the network 150. The video acquisition unit 111 acquires driving video captured by the camera 210 from, for example, a plurality of vehicles 200. The video captured by the camera 210 may be stored in a portable storage device in the vehicle 200. In this case, the video acquisition unit 111 may acquire the video from the portable storage device removed from the vehicle 200. The object detection unit 112 detects an object from the acquired video. For example, the object detection unit 112 detects an object in each frame of the acquired video.

[0028] The label list 114 is a list containing objects (their names) to be compressed with high image quality during video compression. The label list 114 includes, for example, important traffic objects such as vehicles, people, bicycles, and traffic lights. In this embodiment, the objects included in the label list may also include objects that can be identified from images, such as weather, roads, and white or yellow lane lines. The object detection unit 112 detects objects included in the label list 114 in each frame. The object detection unit 112 outputs the object detection results to the video compression unit 113. The object detection results include, for each detected object, location information or coordinate information of the area where the object exists. The object detection unit 112 corresponds to the object detection unit 12 shown in FIG. 1.

[0029] The video compression unit 113 compresses the video acquired by the video acquisition unit 111 using the object detection results. In compressing the video, the video compression unit 113 sets the area of ​​the object detected by the object detection unit 112 as a gaze area. The video compression unit 113 compresses the video so that the image quality of the gaze area is higher than the image quality of areas other than the gaze area. As a result, in the compressed video, the objects included in the label list 114 have relatively high image quality, while the other areas have low image quality. This compression method makes it possible to reduce the file size of the video to about 1 / 10 of the size without compression. The video compression unit 113 stores the compressed video in the video storage unit 130. The video compression unit 113 corresponds to the video compression unit 13 shown in FIG. 1.

[0030] The labeling unit 115 performs a labeling process on the compressed video. The labeling unit 115 assigns a label to each frame by, for example, analyzing the compressed video. The labeling unit 115 inputs the compressed video into, for example, a Vision Language Model (VLM) and assigns a label to each frame of the compressed video using the VLM. The labeling unit 115 stores the assigned labels in the video storage unit 130. In addition to storing the labels, the labeling unit 115 may also store information generated by the VLM in the video storage unit 130 as additional information. The labels and additional information stored in the video storage unit 130 are referenced during video search in the video search device 140.

[0031] The search history acquisition unit 116 acquires a history of search keywords for video searches in the video search device 140. The list update unit 117 updates the label list 114 based on the search keyword history acquired by the search history acquisition unit 116. For example, the list update unit 117 updates the label list based on the number of times or frequency of appearance of each object in the search keyword history. Specifically, the list update unit 117 extracts the names of objects included in each search keyword from the search keyword history and tallies the number of times each extracted object appears. The list update unit 117 may update the label list 114 by adding objects with a high number of tallied appearances, i.e., objects with a high appearance frequency, to the label list 114.

[0032] The list update unit 117 may generate a frequency ranking of objects in the search keyword history and update the label list based on the generated frequency ranking. The frequency ranking may refer to a list obtained by sorting objects extracted from the search keyword in descending order of frequency of appearance. The list update unit 117 may add a predetermined number of objects ranked at the top of the frequency ranking to the label list 114 under predetermined conditions. For example, the list update unit 117 adds to the label list 114 objects that are within a predetermined number of the frequency ranking, for example, within the top 20, and that are not currently included in the label list 114.

[0033] The list updating unit 117 may add objects that are ranked high in the quality ranking to the label list 114, within a range that does not exceed a predetermined upper limit of the number of objects included in the label list 114. Furthermore, the list updating unit 117 may delete from the label list 114, among the objects included in the label list 114, objects that are not included in the frequency ranking or objects that are ranked low in the frequency ranking. The list updating unit 117 may have an exclusion list that specifies objects to be excluded from deletion. The list updating unit 117 does not delete an object included in the exclusion list from the label list 114 even if the object is ranked low in the frequency ranking.

[0034] The list update unit 117 may add to the label list 114 an object related to an object ranked high in the frequency ranking. For example, the list update unit 117 has information indicating the relevance between objects and uses that information to identify an object related to an object ranked high in the frequency ranking. The list update unit 117 may add to the label list 114, in addition to the object ranked high in the frequency ranking, an object related to that object. Specifically, for example, assume that "dog" is ranked high in the frequency ranking. In that case, the list update unit 117 may identify "cat" as an object related to "dog" and add "dog" and "cat" to the label list 114.

[0035] The list update unit 117 may add to the label list 114 objects that are synonyms of objects that rank highly in the frequency ranking. For example, the list update unit 117 has information that defines synonyms of objects, and uses that information to identify objects that are synonyms of objects that rank highly in the frequency ranking. The list update unit 117 may add to the label list 114 objects that are synonyms of objects that rank highly in the frequency ranking, in addition to the objects that rank highly in the frequency ranking. Specifically, for example, assume that "bike" is ranked highly in the frequency ranking. In that case, the list update unit 117 may identify "motorcycle" as a synonym of "bike" and add "motorcycle" and "motorcycle" to the label list 114. The list update unit 117 corresponds to the list update unit 14 shown in FIG. 1 .

[0036] Next, the operation procedure will be described. Fig. 5 is a flowchart showing the operation procedure of the image processing device 110. The operation procedure of the image processing device 110 corresponds to an image processing method. The video acquisition unit 111 acquires video captured by the camera 210 (step S1). The object detection unit 112 detects objects included in the label list 114 in the video acquired in step S1 (step S2).

[0037] The video compression unit 113 compresses the video acquired in step S1 using the result of object detection in step S2 (step S3). In step S3, the video compression unit 113 compresses the video so that the image quality of the area containing the object detected in step S2 is higher than the image quality of the other areas. The label assignment unit 115 performs a labeling process on the compressed video (step S4). The video compression unit 113 stores the compressed video in the video storage unit 130, and the label assignment unit 115 stores the assigned labels in the video storage unit 130 (step S5). The video stored in the video storage unit 130 is used for video search in the video search device 140.

[0038] The search history acquisition unit 116 acquires a search keyword history from the video search device 140. The list update unit 117 updates the label list 114 based on the acquired search keyword history (step S6). In step S6, the list update unit 117 adds, for example, objects that are frequently searched for by users to the label list 114. Note that step S6 does not necessarily have to be performed after steps S1 to S5. Step S6 may be repeatedly performed at any interval, independent of the operations of steps S1 to S5.

[0039] In this embodiment, the object detection unit 112 detects an object included in the label list 114. The video compression unit 113 sets the area of ​​the detected object as a focus area and compresses the video so that the image quality of the focus area is relatively higher than the image quality of areas other than the focus area. By performing this type of video compression, the image processing device 110 can store video with a reduced overall data volume in the video storage unit 130 while maintaining high image quality in important areas of the video. Therefore, this embodiment can reduce storage costs when saving video.

[0040] In this embodiment, the list update unit 117 updates the label list 114 based on the search keyword history for video searches performed by the video search device 140. Objects that are frequently searched for in video searches are considered to be objects for which the user has a high need. The list update unit 117, for example, tallies the number of times an object appears in the search keyword history and generates a ranking of the objects that frequently appear in video searches. The list update unit 117 adds, for example, objects that rank highly in the frequency ranking, i.e., frequently searched objects, to the label list 114. After updating the label list 114, the video compression unit 113 compresses the video so that, for example, the area of ​​the object added to the label list 114 has relatively high image quality. This allows the user to obtain videos containing the desired object in high image quality during video searches, thereby realizing video searches that meet the needs of the user who is the searcher.

[0041] In the above embodiment, the image processing device 110 does not necessarily need to be configured as a single physical device. In the present disclosure, the image processing device 110 may be configured using multiple physically separated devices. For example, the image processing device 110 may be separated into a first device having an image acquisition unit 111, an object detection unit 112, an image compression unit 113, and a label list 114, a second device having a label assignment unit 115, and a third device having a search history acquisition unit 116 and a list update unit 117. In this case, the first device may be installed in each vehicle 200. The first device, the second device, and the third device may be connected via a network 150. When the first device is installed in each vehicle 200, the first device compresses video in each vehicle 200 and transmits the compressed video to the second device via the network 150. In this case, traffic on the network 150 can be reduced compared to when video is transmitted without compression.

[0042] In the present disclosure, the image processing device 110 and the video search device 140 may each be configured as a computer device or a server device. Fig. 6 is a block diagram showing an example configuration of a computer device that can be used as the image processing device 110 or the video search device 140. The computer device 500 includes a processor 510 such as a central processing unit (CPU), a storage unit 520, a read only memory (ROM) 530, a random access memory (RAM) 540, a communication interface (IF) 550, and a user interface 560.

[0043] The communication interface 550 is an interface for connecting the computer device 500 to a communication network via wired communication means, wireless communication means, etc. The user interface 560 includes a display unit such as a display, and an input unit such as a keyboard, a mouse, and a touch panel.

[0044] The storage unit 520 is an auxiliary storage device that can store various types of data. The storage unit 520 does not necessarily have to be a part of the computer device 500, but may be an external storage device or cloud storage connected to the computer device 500 via a network.

[0045] The ROM 530 is a non-volatile storage device. For example, a semiconductor storage device with a relatively small capacity, such as a flash memory, is used for the ROM 530. The programs executed by the CPU 510 can be stored in the storage unit 520 or the ROM 530. The storage unit 520 or the ROM 530 stores various programs for realizing the functions of each unit in the image processing device 110, for example.

[0046] The program includes instructions (or software code) that, when loaded into a computer, cause the computer to perform one or more functions described in the embodiments. The program may be stored in a non-transitory computer-readable medium or a tangible storage medium. By way of example and not limitation, computer-readable media or tangible storage media include RAM, ROM, flash memory, SSD or other memory technologies, Compact Disc (CD), digital versatile disc (DVD), Blu-ray (registered trademark) disc or other optical disk storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage devices. The program may also be transmitted on a transitory computer-readable medium or communication medium. By way of example and not limitation, transitory computer-readable media or communication media include electrical, optical, acoustic, or other forms of propagated signals.

[0047] The RAM 540 is a volatile storage device. Various semiconductor memory devices such as dynamic random access memory (DRAM) or static random access memory (SRAM) are used for the RAM 540. The RAM 540 can be used as an internal buffer for temporarily storing data, etc. The CPU 510 loads a program stored in the storage unit 520 or the ROM 530 into the RAM 540 and executes it. The CPU 510 executes the program, thereby realizing the functions of each unit in the image processing device 110. The CPU 510 may have an internal buffer for temporarily storing data, etc.

[0048] Although the embodiments of the present disclosure have been described in detail above, the present disclosure is not limited to the above-described embodiments, and changes and modifications to the above-described embodiments that do not deviate from the spirit of the present disclosure are also included in the present disclosure.

[0049] Each drawing is merely an example for describing one or more embodiments. Each drawing may not relate to only one particular embodiment, but may also relate to one or more other embodiments. As will be understood by those skilled in the art, various features or steps described with reference to any one drawing can be combined with features or steps shown in one or more other drawings to create, for example, an embodiment not explicitly shown or described. Not all features or steps shown in any one drawing are necessary to describe an exemplary embodiment, and some features or steps may be omitted. The order of steps described in any drawing may be changed as appropriate.

[0050] For example, some or all of the above embodiments can be described as, but are not limited to, the following supplementary notes.

[0051] [Supplementary Note 1] An image processing device comprising: an object detection unit that detects from a video an object included in a label list that includes one or more objects; a video compression unit that compresses the video so that the image quality of a region of the video in which the detected object is located is higher than the image quality of a region of the video other than the region of the detected object; and a list update unit that updates the label list based on a history of search keywords specified by a user in a video search that searches the video using search keywords.

[0052] [Supplementary Note 2] The image processing device according to Supplementary Note 1, wherein the list update unit extracts an object included in the search keyword, and updates the label list based on an appearance frequency of the object in a history of the search keyword.

[0053] [Supplementary Note 3] The image processing device according to Supplementary Note 1 or 2, wherein the list update unit generates a ranking of frequent appearances of objects in the history of the search keywords, and updates the label list based on the ranking of frequent appearances.

[0054] [Supplementary Note 4] The image processing device according to Supplementary Note 3, wherein the list update unit adds a predetermined number of objects that are ranked high in the frequency ranking to the label list under a predetermined condition.

[0055] [Supplementary Note 5] The image processing device according to any one of Supplementary Notes 1 to 4, wherein the video compression unit stores the compressed video in a video storage unit.

[0056] [Supplementary Note 6] The image processing device according to Supplementary Note 5, wherein the video stored in the video storage unit is searched for in the video search.

[0057] [Supplementary Note 7] The image processing device according to Supplementary Note 5 or 6, further comprising a labeling unit that assigns a label to the compressed video by analyzing the compressed video and stores the label assigned to the compressed video in the video storage unit.

[0058] [Supplementary Note 8] The image processing device according to Supplementary Note 7, wherein the labeling unit inputs the compressed video into a Vision Language Model (VLM) and assigns a label to the compressed video using the VLM.

[0059] [Supplementary Note 9] A video search system comprising: an image processing device according to any one of Supplementary Notes 1 to 8; and a video search device that searches for video related to a search keyword specified by a user from a video storage unit that stores the compressed video, based on the search keyword.

[0060] [Supplementary Note 10] An image processing method comprising: detecting, from a video, an object included in a label list including one or more objects; compressing the video so that the image quality of a region of the video where the detected object is located is higher than the image quality of a region of the video other than the region of the detected object; and updating the label list based on a history of search keywords designated by a user in a video search in which the video is searched using search keywords.

[0061] [Supplementary Note 11] A program that causes a computer to execute a process of: detecting, from a video, an object included in a label list that includes one or more objects; compressing the video so that the image quality of an area in the video where the detected object is located is higher than the image quality of an area in the video other than the area in which the detected object is located; and updating the label list based on a history of search keywords specified by a user in a video search that searches the video using search keywords.

[0062] Some or all of the elements (e.g., configurations and functions) described in Supplementary Notes 2 to 8 that are dependent on Supplementary Note 1 may also be dependent on Supplementary Notes 10 and 11 in the same dependency relationship as Supplementary Notes 2 to 8. Some or all of the elements described in any Supplementary Note may be applied to various hardware, software, recording means for recording software, systems, and methods.

[0063] 10: Video retrieval system 11: Image processing device 12: Object detection unit 13: Video compression unit 14: List update unit 20: Video storage unit 30: Video retrieval device 100: Video retrieval system 110: Image processing device 111: Video acquisition unit 112: Object detection unit 113: Video compression unit 114: Label list 115: Label assignment unit 116: Search history acquisition unit 117: List update unit 14: List update unit 130: Video storage unit 140: Video retrieval device 141: Search keyword input unit 142: Search unit 150: Network 200: Vehicle 210: Camera 500: Computer device 510: Processor 520: Storage unit 530: ROM 540: RAM 550: Communication IF 560: User IF

Claims

1. An image processing device comprising: an object detection unit that detects from a video an object included in a label list that includes one or more objects; a video compression unit that compresses the video so that the image quality of an area in the video where the detected object is located is higher than the image quality of areas in the video other than the area where the detected object is located; and a list update unit that updates the label list based on a history of search keywords specified by a user in a video search that searches the video using search keywords.

2. The image processing device according to claim 1, wherein the list update unit extracts objects included in the search keyword and updates the label list based on the frequency of appearance of the objects in the history of the search keyword.

3. The image processing device according to claim 1 or 2, wherein the list update unit generates a ranking of objects that frequently appear in the history of the search keywords, and updates the label list based on the ranking of objects that frequently appear.

4. The image processing device according to claim 3, wherein the list update unit adds a predetermined number of objects that are ranked high in the frequency ranking to the label list under a predetermined condition.

5. The image processing device according to claim 1 or 2, wherein the video compression unit stores the compressed video in a video storage unit.

6. The image processing device according to claim 5, wherein the video stored in the video storage unit is searched for in the video search.

7. The image processing device according to claim 5, further comprising a labeling unit that assigns a label to the compressed video by analyzing the compressed video and stores the label assigned to the compressed video in the video storage unit.

8. The image processing device according to claim 7, wherein the labeling unit inputs the compressed video into a Vision Language Model (VLM) and assigns labels to the compressed video using the VLM.

9. A video search system comprising: an image processing device according to claim 1 or 2; and a video search device that searches a video storage unit that stores the compressed video for video related to the search keyword, based on the search keyword specified by a user.

10. An image processing method that detects from a video an object included in a label list that includes one or more objects, compresses the video so that the image quality of the area of ​​the detected object in the video is higher than the image quality of areas other than the area of ​​the detected object in the video, and updates the label list based on the history of search keywords specified by a user in a video search that searches the video using search keywords.

11. A program that causes a computer to execute a process of detecting an object included in a label list containing one or more objects from a video, compressing the video so that the image quality of the area of ​​the video where the detected object is located is higher than the image quality of areas of the video other than the area of ​​the detected object, and updating the label list based on the history of search keywords specified by a user in a video search that searches the video using search keywords.

Images