Server, terminal, video collection system, video collection method, and recording medium

Abstract

Provided are a server, a terminal, a video collection system, a video collection method, and a recording medium with which it is possible to accurately extract only a video showing a desired scene from a large amount of video data.　This server comprises: a reception means that receives, from a terminal, a first video generated by compressing a video captured on the terminal side using a prescribed compression parameter; an estimation means that estimates a scene shown in the first video; and a request means that, when the scene shown in the video matches a prescribed condition, requests the terminal to transmit a second video having a lower compression rate than the first video. The terminal transmits the first video to the server, and transmits the second video in response to a request from the server.

Description

Server, terminal, video collection system, video collection method and recording medium

[0001] The present invention relates to a server, a terminal, a video collection system, a video collection method, and a recording medium.

[0002] Patent Document 1 discloses a drive recorder that can suppress an increase in communication traffic due to the transmission of captured video. According to this document, the drive recorder includes a normal storage unit that stores video acquired by a video acquisition unit as normal video, and a compression storage unit that stores compressed video, which is video obtained by compressing the data volume of the normal video stored in the normal storage unit. When the drive recorder determines that a predetermined transmission condition is met, the drive recorder performs a compression transmission process in which the compressed video for the time including the time when the transmission condition is met is set as transmission compressed video and transmits the transmission compressed video to a predetermined destination.

[0003] International Publication No. 2020 / 044759

[0004] There is a need to accurately extract only videos showing desired scenes from a large amount of video content captured by cameras or the like in remote locations. Although various scene detection techniques are known, extracting only videos showing desired scenes from a large amount of video content requires considerable computational cost and time. In this regard, the drive recorder disclosed in Patent Document 1 only transmits the compressed video from a time including the time when a transmission condition is met.

[0005] The present disclosure aims to provide a server, a terminal, a video collection system, a video collection method, and a recording medium that can accurately extract video that shows a desired scene from a large amount of video data.

[0006] According to a first aspect, a server is provided that includes a receiving means for receiving from a terminal a first image generated by compressing an image captured on the terminal side with a predetermined compression parameter, a scene estimation means for estimating a scene captured in the first image, and a request means for requesting the terminal to transmit a second image having a lower compression rate than the first image if the scene captured in the image meets predetermined conditions.

[0007] According to a second aspect, there is provided a terminal that transmits the first video to the server and transmits the second video in response to a request from the server.

[0008] According to a third aspect, there is provided a video collection system including: a server having a receiving means for receiving from a terminal a first video generated by compressing video taken on the terminal side with predetermined compression parameters; a scene estimation means for estimating a scene captured in the first video; and a request means for requesting the terminal to transmit a second video having a lower compression rate than the first video if the scene captured in the video meets predetermined conditions; and a terminal that transmits the first video to the server and transmits the second video in response to a request from the server.

[0009] According to a fourth aspect, there is provided a video collection method that receives from a terminal a first video generated by compressing video captured on the terminal side with predetermined compression parameters, estimates a scene captured in the first video, and, if the scene captured in the video meets predetermined conditions, requests the terminal to transmit a second video that has a lower compression rate than the first video.

[0010] According to a fifth aspect, there is provided a recording medium having recorded thereon a program that causes a computer to execute the following processes: receiving from a terminal a first image generated by compressing an image captured on the terminal side with predetermined compression parameters; estimating a scene captured in the first image; and, if the scene captured in the image meets predetermined conditions, requesting the terminal to transmit a second image having a lower compression rate than the first image.

[0011] According to the present disclosure, it is possible to provide a server, terminal, video collection system, video collection method, and recording medium that can accurately extract video that shows a desired scene from a large amount of video data.

[0012] FIG. 1 is a diagram showing one configuration of the present disclosure. FIG. 2 is a flow chart showing the operation of the present disclosure. FIG. 3 is a diagram for explaining the operation of the present disclosure. FIG. 4 is a diagram showing one configuration of the present disclosure. FIG. 5 is a functional block diagram showing the configuration of an in-vehicle terminal of the present disclosure. FIG. 6 is a functional block diagram showing the configuration of a server of the present disclosure. FIG. 7 is a sequence diagram for explaining the operation of the present disclosure. FIG. 8 is a functional block diagram showing another configuration of a server of the present disclosure. FIG. 9 is a sequence diagram for explaining the operation of the present disclosure. FIG. 10 is a functional block diagram showing another example configuration of a server of the present disclosure. FIG. 11 is a sequence diagram for explaining the operation of the present disclosure. FIG. 12 is a diagram showing the configuration of a computer constituting a server or an in-vehicle terminal of the present disclosure.

[0013] First, an overview of one embodiment of the present disclosure will be described with reference to the drawings. In this disclosure, the drawings relate to one or more embodiments. The reference numerals in the drawings attached to this overview are attached to each element for convenience as an example to facilitate understanding, and are not intended to limit the present disclosure to the illustrated form. Furthermore, connecting lines between blocks in the drawings and the like referred to in the following description include both bidirectional and unidirectional lines. Unidirectional arrows are used to schematically indicate the flow of main signals (data) and do not exclude bidirectionality. A program is executed via a computer device, which includes, for example, a processor, a storage device, an input device, a communication interface, and, if necessary, a display device. Furthermore, this computer device is configured to be able to communicate with internal or external devices (including computers) via the communication interface, whether wired or wireless. Although ports or interfaces are present at the input / output connection points of each block in the drawings, they are not shown.

[0014] In one embodiment, the present disclosure can be realized by a server 10 including a receiving means 11, a scene estimation means 12, and a request means 13, as shown in FIG. 1 . More specifically, the receiving means 11 receives, from a terminal 20, a first video generated by compressing a video captured by a camera C or the like on the terminal 20 using predetermined compression parameters. The scene estimation means 12 estimates a scene captured in the first video. If the scene captured in the video meets predetermined conditions, the request means 13 requests the terminal 20 to transmit a second video having a lower compression rate than the first video. The second video may be the video before being compressed by the terminal 20.

[0015] The server 10 configured as described above operates as follows: First, the server 10 receives from the terminal 20 a first video image generated by compressing a video image captured on the terminal side using a predetermined compression parameter (step S01 in FIG. 2).

[0016] Next, the server 10 estimates the scene captured in the first video (step S02 in FIG. 2), and determines whether the scene captured in the first video matches a predetermined condition (step S03 in FIG. 2).

[0017] If it is determined that the scene in the video matches a predetermined condition, the server 10 requests the terminal 20 to transmit a second video having a lower compression rate than the first video (step S04 in FIG. 2). Next, the server 10 receives the second video from the terminal 20 (step S05 in FIG. 2).

[0018] Furthermore, if it is determined that the first video does not match the search criteria (No in step S03 in FIG. 2), the server 10 does not request the terminal 20 to transmit the second video, thereby preventing unnecessary data from being transmitted.

[0019] FIG. 3 is a diagram illustrating the operation of the present disclosure. The terminal 20 transmits to the server 10 a first video 21 obtained by compressing a video 23 captured on the terminal side using a compression process 22. The server 10 uses the scene estimation means 12 to perform scene determination on the compressed first video and determine whether it is a necessary video. If the determination results in a necessary video, the server 10 requests the terminal 20 to transmit a second video with a larger amount of information. By repeating the above process, the server 10 can collect videos 14 of desired scenes with a smaller amount of communication traffic. For example, assume there are 100 videos, each 100 MB in size. If these videos are transmitted without compression, the communication traffic would be 10,000 MB, or approximately 10 GB. In contrast, if the method of the present disclosure is used to compress the 100 MB video to 1 / 10 and then transmit it, the communication traffic would be 1,000 MB, or approximately 1 GB. Even if there are 10 videos that contain the desired scene, the total communication volume is 100 MB x 10 = 1 GB, and the overall communication volume can be reduced to 2 GB. Furthermore, the method disclosed herein has the advantage of being able to significantly reduce the calculation cost and processing time in the scene estimation means 12.

[0020] As described above, according to the present disclosure, it is possible to accurately extract only the video showing a desired scene from a large amount of video data. Note that, although one terminal 20 is shown in the example of Fig. 1, there may be two or more terminals 20 transmitting video to the server 10. Since each terminal 20 compresses the video (first video) before transmitting it to the server 10, the load on the network can also be reduced.

[0021] [First Embodiment] Next, a first embodiment will be described that enables collection of videos showing desired scenes from a large amount of video data captured by a vehicle or the like equipped with a camera or the like. FIG. 4 is a diagram showing one configuration of the present disclosure. Referring to FIG. 4, an in-vehicle terminal 200 capable of transmitting videos captured by a camera C to a server 100 is shown. The in-vehicle terminal 200 transmits the videos captured by the camera C mounted on the vehicle to the server 100. The server 100 collects videos showing desired scenes from the videos received from the in-vehicle terminal 200.

[0022] Fig. 5 is a functional block diagram showing the configuration of the in-vehicle terminal 200 of the present disclosure. Referring to Fig. 5, the configuration includes a compression unit 201, a transmission unit 202, a transmission request processing unit 203, and a video storage unit 204. Video captured by camera C is stored in the video storage unit 204 and sent to the compression unit 201. It is desirable that the video captured by camera C has high image quality that meets a certain standard.

[0023] The compression unit 201 reduces the data size of the video captured by the camera C and then outputs the video to the transmission unit 202. Methods for reducing this data size include lowering the resolution, lowering the frame rate, and lowering the image quality (including control of increasing the image quality of important areas in the video and lowering the image quality of other areas). Hereinafter, this reduction in the data size of the video will also be referred to as "compression." In this embodiment, it is assumed that appropriate values ​​are set as parameters for this compression so as not to interfere with scene estimation in the server 100.

[0024] The transmitting unit 202 transmits the video (first video) compressed by the compressing unit 201 to the server 100. Furthermore, when the transmitting unit 202 receives a request from the transmission request processing unit 203, it transmits the video (second video) stored in the video storage unit 204 to the server 100.

[0025] When transmission request processing unit 203 receives a request to transmit video from server 100, it identifies the corresponding video from video storage unit 204 and requests transmission unit 202 to transmit it to server 100. If the video stored in video storage unit 204 has high image quality and is expected to place a load on the network, transmission request processing unit 203 may reduce the data size of the video stored in video storage unit 204.

[0026] 6 is a functional block diagram showing the configuration of the server 100 of the present disclosure. Referring to FIG. 6, the server 100 includes a receiving unit 101, a scene estimation unit 102, a request unit 103, and a video storage unit 104.

[0027] When the receiving unit 101 receives a video (first video) from the in-vehicle terminal 200, it sends the received video to the scene estimation unit 102. Therefore, this receiving unit 101 corresponds to the above-mentioned receiving means 11. Furthermore, when the receiving unit 101 receives a video (second video) sent in response to a transmission request for the second video to the in-vehicle terminal 200, it stores the video in the video storage unit 104.

[0028] The scene estimation unit 102 estimates a scene captured in an input video (first video), determines whether the scene conforms to conditions for identifying a desired scene, and outputs the result to the request unit 103. Various scene estimation technologies can be used for this scene estimation. Furthermore, the scene estimation unit 102 can also use a large-scale language model or a VLM (Vision-Language Model) capable of performing scene analysis of video. Known examples of VLMs capable of scene analysis include LLaVA (Large Language and Vision Assistant). When the scene estimation unit 102 is configured with an LLM or VLM capable of handling video, the conditions may be in a natural language format. In this case, by inputting the first video to the scene estimation unit 102 together with conditions written in a natural language format, a determination result as to whether the input video (first video) conforms to the conditions can be obtained. This scene estimation unit 102 corresponds to the scene estimation means 12 described above.

[0029] When the scene estimation unit 102 determines that the input video (first video) contains a desired scene, the request unit 103 requests the in-vehicle terminal 200 to transmit a video (second video) with a lower compression rate than the video (first video) for which the scene estimation was performed. This request unit 103 corresponds to the request means 13 described above.

[0030] The video storage unit 104 stores the video (second video) with a low compression rate received from the in-vehicle terminal 200. The video stored in the video storage unit 104 has high image quality and is a large amount of video for each scene, and can be suitably used, for example, as training data for machine learning for developing an image analysis engine for autonomous driving.

[0031] Next, the operation of this embodiment will be described in detail with reference to the drawings. Fig. 7 is a sequence diagram for explaining the operation of the present disclosure. Referring to Fig. 7, first, the in-vehicle terminal 200 transmits the compressed first video to the server 100 (step S002).

[0032] The server 100, which has received the first video, performs scene estimation for the received first video (step S002). Next, the server 100 determines whether the estimated scene satisfies a pre-set condition (step S003). For example, if a condition such as "Save a video of a car and a bicycle traveling side by side" is set in natural language format, the server 100 determines whether the estimated scene satisfies this condition.

[0033] If the result of the above judgment is that the first video contains the desired scene that meets the conditions, the server 100 requests the in-vehicle terminal 200 to send a second video of better image quality (lower compression rate) than the first video (step S004).

[0034] Upon receiving the request, the in-vehicle terminal 200 transmits the second video to the server 100 (step S005).

[0035] The server 100 receives the second video from the in-vehicle terminal 200 and stores the received second video (step S006).

[0036] FIG. 8 is a diagram illustrating the operation of the present disclosure. As shown in FIG. 8, the in-vehicle terminal 200 transmits compressed video (first video) to the server 100. The server 100 performs scene estimation on this compressed video (first video). This has the advantage of significantly reducing calculation costs and time while maintaining the accuracy of scene estimation compared to analyzing high-quality video. If the scene estimation results indicate that a higher-quality video is required, the server 100 requests the in-vehicle terminal 200 to transmit a second video with better image quality (lower compression rate). The in-vehicle terminal 200 then transmits the second video to the server 100.

[0037] As described above, according to this embodiment, it is possible to accurately extract only the video containing the desired scene from the large amount of video data sent from the in-vehicle terminal 200. Furthermore, in the above-described embodiment, the in-vehicle terminal 200 transmits the video before compression by the in-vehicle terminal 200 as the second video. However, the second video may be transmitted after recompression using predetermined compression parameters. [Second Embodiment] Next, a second embodiment will be described in which a compression parameter optimization function is provided in the server. FIG. 9 is a diagram showing another configuration of the server 100a of the present disclosure. The differences from the server 100 of the first embodiment shown in FIG. 6 are the addition of a compression parameter determination unit 105 and changes to the operation of the scene estimation unit 102a and the request unit 103a. Since the other configurations are substantially the same as those of the first embodiment, the following description will focus on the operational differences.

[0038] If the parameters (compression parameters) used when generating the first video are fixed, the scene estimation capability of the scene estimation unit 102 may vary depending on the brightness of the image, the speed of the subject, and so on.

[0039] When the scene estimation unit 102a of the server 100a performs scene estimation, it calculates the accuracy of the estimation result (scene estimation accuracy). If the scene estimation accuracy is below a predetermined threshold, the scene estimation unit 102a notifies the compression parameter determination unit 105 of the scene estimation accuracy.

[0040] The compression parameter determination unit 105 determines compression parameters for the video to be requested to be retransmitted from the in-vehicle terminal 200a based on the scene estimation accuracy of the scene estimation unit 102a. At this time, the compression parameter determination unit 105 determines the compression parameters so that the compression ratio is lower than the compression parameters used when the first video was compressed previously.

[0041] The request unit 103 a requests the in-vehicle terminal 200 a to transmit the first video compressed using the compression parameters determined by the compression parameter determination unit 105 .

[0042] Figure 10 is a diagram showing another configuration of the on-board terminal 200a of the present disclosure. The first difference from the on-board terminal 200 of the first embodiment shown in Figure 5 is that the transmission request processing unit 203a can accept a retransmission request from the server 100a for the first video compressed with specified compression parameters. The second difference from the on-board terminal 200 of the first embodiment shown in Figure 5 is that the compression unit 201a recompresses the transmitted video with specified compression parameters and outputs it to the transmission unit 202. Since the other configurations are substantially the same as those of the first embodiment, the following description will focus on the operational differences.

[0043] When the transmission request processor 203a receives a request from the server 100a to retransmit the first video compressed with the specified compression parameters, it requests the compressor 201a to compress the video with the specified compression parameters.

[0044] Upon receiving the request to recompress the video, the compression unit 201 a recompresses the video using the specified compression parameters and outputs the recompressed video to the transmission unit 202 .

[0045] 11 is a sequence diagram illustrating an operation added in the present disclosure. Steps S001 and S002 in FIG. 11 are the same as those in the first embodiment, and therefore will not be described here. The server 100 then checks whether the accuracy of scene estimation (scene estimation accuracy) is below a predetermined threshold (step S101). If the check results in the scene estimation accuracy being below the predetermined threshold (Yes in step S101), the server 100 requests the in-vehicle terminal 200a to transmit the first video with modified compression parameters (step S102).

[0046] When the in-vehicle terminal 200a receives a request from the server 100a to retransmit the first video compressed with the specified compression parameters, the in-vehicle terminal 200a recompresses the video and transmits the compressed video to the server 100a (step S103). The subsequent operations are the same as those in the first embodiment, and therefore will not be described again.

[0047] As described above, the server 100a requests the in-vehicle terminal 200a to send video (first video) with modified compression parameters so that the compression rate is lower than that of the received first video. This allows the accuracy of scene estimation for the retransmitted first video to be improved compared to the previous time. Therefore, according to this embodiment, it is possible to recover from a failure in scene estimation due to a problem with the compression parameters or the state of the first video.

[0048] In addition, in the above embodiment, the in-vehicle terminal 200 is described as retransmitting the first video compressed with specified compression parameters as the first video, but the first video may also be transmitted as the video before compression (video stored in the video storage unit 204).

[0049] [Third Embodiment] Next, a third embodiment will be described, which realizes a video recompression function based on scene estimation accuracy in a different form. Fig. 12 is a diagram showing another configuration of a server 100b according to the present disclosure. The difference from the server 100a of the second embodiment shown in Fig. 9 is that the compression parameter determination unit 105b holds a model capable of outputting compression parameters for obtaining a desired scene estimation accuracy for input video. Such a compression parameter determination unit 105b can also be configured using a generation AI capable of handling video.

[0050] When the compression parameter determination unit 105b receives notification of the scene estimation accuracy from the scene estimation unit 102b, it additionally learns the relationship between the compression parameters of the first video for which scene estimation has been performed and the scene estimation accuracy, and updates the model.

[0051] The request unit 103b requests the in-vehicle terminal 200a to transmit the first video compressed using the compression parameters determined by the compression parameter determination unit 105b.

[0052] FIG. 13 is a sequence diagram illustrating the operation of the present disclosure. Steps S001 and S002 in FIG. 13 are the same as those in the first embodiment, and therefore will not be described here. The server 100b then checks whether the accuracy of the scene estimation (scene estimation accuracy) is below a predetermined threshold (step S201). If the check results in the scene estimation accuracy being below the predetermined threshold (Yes in step S201), the server 100b instructs the in-vehicle terminal 200a to retrain using the first image and update the model. The server 100b then recalculates compression parameters using the updated model (step S202). The server 100b then requests the in-vehicle terminal 200a to transmit the first video, specifying the recalculated compression parameters (step S203).

[0053] When the in-vehicle terminal 200a receives a request from the server 100b to retransmit the first video compressed with the specified compression parameters, the in-vehicle terminal 200a recompresses the video and transmits the compressed video to the server 100 (step S204). The subsequent operations are the same as those in the first embodiment, and therefore will not be described again.

[0054] According to this embodiment, in addition to the effects of the second embodiment, the compression parameter determination unit 105b adjusts the compression parameters by receiving feedback on the scene estimation accuracy of the scene estimation unit 102c, which has the advantage of being able to improve the accuracy of scene estimation for equivalent video in the scene estimation unit 102c.

[0055] Although the embodiments of the present disclosure have been described above, the present disclosure is not limited to the above-described embodiments, and further modifications, substitutions, and adjustments can be made without departing from the basic technical concept of the present disclosure. For example, the network configurations, element configurations, and data representation formats shown in the drawings are examples intended to aid in understanding the present disclosure, and are not limited to the configurations shown in these drawings.

[0056] For example, in the above-described first to third embodiments, an example was given in which the video transmitted from the in-vehicle terminal 200, 200a is the subject of scene estimation and storage, but the application of the present disclosure is not limited to this.

[0057] (UC1) Scene collection at construction sites, etc. For example, video captured by a camera installed at a construction site can be transmitted to a terminal. Then, by using the servers 100-100b of the present disclosure, it is possible to extract from a large volume of video footage video that matches a search condition, such as "a worker working at a height without wearing a safety harness." By using such video footage as training data, a system can be constructed that provides appropriate warnings to workers working at a height. (UC2) Scene collection at production lines, etc. For example, video captured by a work status monitoring camera installed on a production line can be transmitted to a terminal. Then, by using the servers 100-100b of the present disclosure, it is possible to extract from a large volume of video footage video that matches a search condition, such as "a scene where soldering failed during the production of product A." By using such video footage as training data, it is possible to construct a system that provides appropriate warnings when a situation that leads to a failure in the soldering process of product A occurs. (UC3) Scene collection at retail stores, etc. For example, video captured by a surveillance camera installed in a retail store can be transmitted to a terminal. Using the server of the present disclosure, it becomes possible to extract from a large amount of video footage videos that meet search criteria such as "scenes showing a man around XX years old holding product B." By using such videos as learning data, it is possible to build a system that analyzes the behavior of the target demographic for each product.

[0058] Furthermore, the images that can be collected by the server of the present disclosure are not limited to those captured by a camera in the visible light range. For example, images captured by an infrared video camera can be transmitted to the terminal for scene estimation and storage. Similarly, images of point cloud data generated by a 3D scanner or images of synthetic aperture radar (SAR) obtained by a synthetic aperture radar can also be transmitted to the terminal for scene estimation and storage.

[0059] (Hardware Configuration) In each embodiment of the present disclosure, each component of each device represents a functional unit block. Some or all of the components of each device are realized by an arbitrary combination of an information processing device 900 and a program, for example, as shown in FIG. 14. FIG. 14 is a block diagram showing an example of the hardware configuration of the information processing device 900 that realizes each component of each device. The information processing device 900 includes, as an example, the following configuration: - CPU (Central Processing Unit) 901 - ROM (Read Only Memory) 902 - RAM (Random Access Memory) 903 - Program 904 loaded into RAM 903 - Storage device 905 that stores the program 904 - Drive device 907 that reads and writes to a recording medium 906 - Communication interface 908 that connects to a communication network 909 - Input / output interface 910 that inputs and outputs data - Bus 911 that connects each component

[0060] Each component of each device in each embodiment is realized by the CPU 901 acquiring and executing a program 904 that realizes the function. That is, the CPU 901 in FIG. 14 executes a scene estimation program and a determination program, and performs an update process for each calculation parameter stored in the RAM 903, the storage device 905, etc. The program 904 that realizes the function of each component of each device is stored in the storage device 905 or the ROM 902 in advance, for example, and is read by the CPU 901 as needed. The program 904 may be supplied to the CPU 901 via the communication network 909, or may be stored in advance on the recording medium 906, and the drive device 907 may read the program and supply it to the CPU 901.

[0061] There are various variations in the method of realizing each device. For example, each device may be realized by any combination of a separate information processing device 900 and a program for each component. Furthermore, multiple components of each device may be realized by any combination of a single information processing device 900 and a program. In other words, each unit (processing means, function) of the above-mentioned server and terminal can be realized by a computer program that causes a processor installed in the device to execute each of the above-mentioned processes using its hardware.

[0062] In addition, some or all of the components of each device may be realized by other general-purpose or dedicated circuits, processors, etc., or a combination of these. These may be configured by a single chip, or by multiple chips connected via a bus.

[0063] Some or all of the components of each device may be realized by a combination of the above-mentioned circuits and programs.

[0064] When some or all of the components of each device are realized by multiple information processing devices, circuits, etc., the multiple information processing devices, circuits, etc. may be centrally or decentralized. For example, the information processing devices, circuits, etc. may be realized as a client-server system, a cloud computing system, or the like, in a form in which each device is connected via a communication network.

[0065] It should be noted that the above-described embodiments are preferred embodiments of the present disclosure, and the scope of the present disclosure is not limited to only the above-described embodiments. In other words, those skilled in the art can modify or substitute the above-described embodiments to construct various modified forms without departing from the gist of the present disclosure.

[0066] A part or all of the above-described embodiments can be described as, but not limited to, the following supplementary notes.

[0067] [Supplementary Note 1] A server comprising: receiving means for receiving from a terminal a first video generated by compressing a video captured on the terminal side with predetermined compression parameters; scene estimation means for estimating a scene captured in the first video; and request means for requesting the terminal to transmit a second video having a lower compression rate than the first video if the scene captured in the video meets predetermined conditions. [Supplementary Note 2] In the above server, the second video may be video before being compressed with the predetermined compression parameters or video compressed with compression parameters different from the predetermined compression parameters. [Supplementary Note 3] The above server may be configured to request the terminal to transmit the video before being compressed with the predetermined compression parameters or video compressed with compression parameters different from the predetermined compression parameters if the scene estimation means cannot estimate the scene captured in the first video. [Supplementary Note 4] The above-mentioned server can be configured to, when the scene estimation means is unable to estimate the scene captured in the first video, additionally learn the relationship between the compression parameters and scene estimation accuracy when the scene could not be estimated, and update a model for selecting the compression parameters, and request the terminal to transmit video compressed with the compression parameters obtained using the updated model. [Supplementary Note 5] In the above-mentioned server, the predetermined condition is that the scene to be collected is captured, and scene-specific videos to be used for machine learning are collected from the video collected by the terminal via the terminal. [Supplementary Note 6] A terminal that transmits the first video to the above-mentioned server and transmits the second video in response to a request from the server. [Supplementary Note 7] A video collection system including: a server comprising: receiving means for receiving from a terminal a first video generated by compressing a video taken on the terminal side with a predetermined compression parameter; estimation means for estimating a scene captured in the first video; and request means for requesting the terminal to transmit a second video having a lower compression rate than the first video when the scene captured in the video meets a predetermined condition; and a terminal that transmits the first video to the server and transmits the second video in response to a request from the server.[Supplementary Note 8] A video collection method comprising: receiving, from a terminal, a first video generated by compressing a video captured by the terminal with predetermined compression parameters; estimating a scene captured in the first video; and, if the scene captured in the video meets predetermined conditions, requesting the terminal to transmit a second video with a compression rate lower than that of the first video. [Supplementary Note 9] A recording medium having recorded thereon a program causing a computer to execute the following steps: receiving, from a terminal, a first video generated by compressing a video captured by the terminal with predetermined compression parameters; estimating a scene captured in the first video; and, if the scene captured in the video meets predetermined conditions, requesting the terminal to transmit a second video with a compression rate lower than that of the first video. Note that the embodiments described in the above supplementary notes can be combined with each other after making necessary modifications. For example, a configuration that combines the contents of Supplementary Note 2 and Supplementary Note 3 is also within the scope of the disclosure of this specification. In this case, the server transmits the video before being compressed with the predetermined compression parameters as the second video, and requests retransmission of the first video with the changed compression parameters. Note that the above-mentioned Supplementary Notes 7 to 9 can be expanded into the forms of Supplementary Notes 2 to 5, similar to Supplementary Note 1.

[0068] The disclosures of the above-cited patent documents are incorporated herein by reference and may be used as the basis or part of this disclosure, as necessary. Modifications and adjustments of the embodiments and examples are possible within the scope of this disclosure (including the claims), and further based on its basic technical concept. Furthermore, various combinations and selections (including partial deletions) of various disclosed elements (including elements of each claim, each element of each embodiment or example, each element of each drawing, etc.) are possible within the scope of this disclosure. In other words, this disclosure naturally includes various modifications and alterations that would be possible by a person skilled in the art in accordance with the entire disclosure, including the claims, and the technical concept. In particular, with regard to the numerical ranges described herein, any numerical value or subrange within that range should be construed as specifically described, even if not otherwise specified. Furthermore, the disclosures of the above-cited documents, when used in part or in whole in combination with the disclosures herein as part of this disclosure, in accordance with the spirit of this disclosure, are also deemed to be included in the disclosures of this application.

[0069] DESCRIPTION OF SYMBOLS 10, 100, 100a, 100b Server 11 Receiving means 12 Estimating means 13 Requesting means 20 Terminal 21 First video 22 Compression processing 23, 14 Video 101 Receiving unit 102 Scene estimation unit 103 Requesting unit 104 Video storage unit 200 In-vehicle terminal 201 Compressing unit 202 Transmitting unit 203 Transmission request processing unit 204 Video storage unit 900 Information processing device 901 CPU (Central Processing Unit) 902 ROM (Read Only Memory) 903 RAM (Random Access Memory) 904 Program 905 Storage device 906 Recording medium 907 Drive device 908 Communication interface 909 Communication network 910 Input / output interface 911 Bus C Camera

Claims

1. A server comprising: a receiving means for receiving from a terminal a first image generated by compressing an image captured on the terminal side using predetermined compression parameters; a scene estimation means for estimating a scene captured in the first image; and a request means for requesting the terminal to transmit a second image having a lower compression rate than the first image when the scene captured in the image meets predetermined conditions.

2. The server according to claim 1, wherein the second video is video before being compressed with the predetermined compression parameters or video compressed with compression parameters different from the predetermined compression parameters.

3. A server according to claim 1 or 2, which requests the terminal to transmit the video before being compressed with the predetermined compression parameters or the video compressed with compression parameters different from the predetermined compression parameters, if the scene estimation means is unable to estimate the scene captured in the first video.

4. A server according to claim 1 or 2, wherein, when the scene estimation means is unable to estimate the scene captured in the first video, the server additionally learns the relationship between the compression parameters and scene estimation accuracy when the scene cannot be estimated, updates the model for selecting the compression parameters, and requests the terminal to transmit video compressed using the compression parameters obtained using the updated model.

5. A server according to any one of claims 1 to 4, wherein the specified condition is that the scene to be collected is included, and the server collects, via the terminal, videos by scene to be used for machine learning from the videos collected by the terminal.

6. A terminal that transmits the first video to a server according to any one of claims 1 to 5, and transmits the second video in response to a request from the server.

7. A video collection system comprising: a server having: receiving means for receiving from a terminal a first video generated by compressing video taken on the terminal side using predetermined compression parameters; estimation means for estimating a scene captured in the first video; and request means for requesting the terminal to transmit a second video having a lower compression rate than the first video when the scene captured in the video meets predetermined conditions; and a terminal that transmits the first video to the server and transmits the second video in response to a request from the server.

8. A video collection method comprising: receiving from a terminal a first video generated by compressing video captured on the terminal side using predetermined compression parameters; estimating a scene captured in the first video; and, if the scene captured in the video meets predetermined conditions, requesting the terminal to transmit a second video having a lower compression rate than the first video.

9. A recording medium having recorded thereon a program that causes a computer to execute the following processes: receiving from a terminal a first image generated by compressing an image captured on the terminal side using predetermined compression parameters; estimating the scene captured in the first image; and, if the scene captured in the image meets predetermined conditions, requesting the terminal to transmit a second image that has a lower compression rate than the first image.

Images