VR video playback control method and system

Abstract

The present application provides a VR video playback control method and system. Since what are determined are the video recording subjects of the target scene and the video recording subjects with defects in the target scene, based on the video recording subjects of the target scene and the video recording subjects with defects in the target scene, the video recording subjects without defects in the target scene are determined to be the distribution in sequence order. In this way, the distribution without defects can be determined more accurately, so that the confidence of the sample data obtained by performing the optimization operation on the subjects without defects is high. In addition, since the video recording subjects without defects in the target scene are determined first, and then the optimization operation is performed on the subjects without defects, a video optimization result is generated according to the optimization operation, and a video playback instruction is determined in combination with the video optimization result. Therefore, accurate and reliable control of video playback can be achieved.

Description

Technical Field

[0001] The present application relates to the field of video control technology, and in particular to a VR video playback control method and system. Background Art

[0002] VR videos can show viewers 360-degree panoramic shots, making them feel as if they are in the scene. This requires VR videos to maintain high continuity during the shooting process and minimize the switching of shots. From the perspective of the lens, VR videos have many requirements. In traditional video shooting, you can switch between various shooting lenses at will according to the video content without any pressure. However, in VR videos, due to the special position of the audience, the method of lens narration will be subject to many restrictions.

[0003] Nowadays, as VR videos are increasingly associated with people's lives, people pay more attention to the video playback effect. In actual operation, when the video data is being played, there may be some defects in the video data, making it difficult to guarantee the video playback effect. Summary of the invention

[0004] In order to improve the technical problems existing in the related technologies, the present application provides a VR video playback control method and system.

[0005] In a first aspect, a VR video playback control method is provided, which is applied to a video playback control system, and the method at least includes: determining a video recording subject of a target scene and a video recording subject with defects in the target scene based on VR video data that needs to be processed; determining a video recording subject without defects in the target scene in combination with the video recording subject of the target scene and the video recording subject with defects in the target scene; performing optimization operations on the video recording subject without defects in combination with the video recording subject without defects in the target scene; generating a video optimization result based on the optimization operation, and determining a video playback instruction in combination with the video optimization result.

[0006] In an independently implemented embodiment, the method of determining the video recording subject of the target scene based on the VR video data to be processed includes: performing a feature extraction operation on the VR video data to be processed to obtain video event features of the target scene; and determining the video recording subject of the target scene in combination with the video event features of the target scene.

[0007] In an independently implemented embodiment, the method of determining the video recording subject having defects in the target scene based on the VR video data that needs to be processed includes: determining the video recording subject having defects in the target scene based on a video classification thread and the VR video data that needs to be processed.

[0008] In an independently implemented embodiment, the method of determining the video recording subject having defects in the target scene based on the video classification thread and the VR video data to be processed includes: determining the video category data including the target scene in combination with the VR video data to be processed; determining the target sample data to be input to the video classification thread in combination with the video category data; processing the target sample data by the video classification thread to obtain output sample data; and determining the video recording subject having defects in the target scene in combination with the output sample data.

[0009] In an independently implemented embodiment, the combining of the VR video data to be processed to determine the video category data including the target scene includes: combining the VR video data to be processed and a mapping list to determine the mapped sample data; performing clustering processing on the mapped sample data to obtain the video category data; the combining of the output sample data to determine the video recording subject with defects in the target scene includes: combining the output sample data and the mapping list to determine the video recording subject with defects in the target scene.

[0010] In an independently implemented embodiment, the output sample data includes: a target list consistent with the number of sequences of the target sample data or the video type data, and an attribute factor of each attribute in the target list, which characterizes the possibility that the sequence corresponding to each attribute belongs to the target scene or the non-target scene; combining the output sample data to determine the video recording subject with defects in the target scene includes: treating the attribute factor of each attribute in the target list as the attribute factor corresponding to each sequence in the video type data; combining the attribute factor corresponding to each sequence in the video type data and the mapping list to determine the video recording subject with defects in the target scene.

[0011] In an independently implemented embodiment, the method of combining the attribute factors corresponding to each sequence in the video type data and the mapping list to determine the video recording subject with defects in the target scene includes: combining the attribute factors corresponding to each sequence in the video type data and the mapping list to determine the attribute factors corresponding to each sequence in the partial sample data including the target scene in the VR video data to be processed that corresponds to the video type data; combining the attribute factors corresponding to each sequence in the partial sample data to determine the video recording subject with defects in the target scene.

[0012] In an independently implemented embodiment, the method of combining the attribute factors corresponding to each sequence in the partial sample data to determine the video recording subject with defects in the target scene includes: determining a target attribute factor that meets a first target value requirement on the video recording subject of the target scene from the attribute factors corresponding to each sequence in the partial sample data; determining a video recording subject with a second set sequence in the VR video data to be processed corresponding to a first set sequence in the partial sample data; and determining the video recording subject with defects in the target scene by combining the video recording subject corresponding to the target attribute factor on the partial sample data and the video recording subject of the second set sequence.

[0013] In an independently implemented embodiment, the method of combining the attribute factors corresponding to each sequence in the partial sample data to determine the video recording subject with defects in the target scene includes: determining the attribute factors corresponding to each sequence in a set range of the VR video data that needs to be processed except the partial sample data; determining, on the video recording subject of the target scene, the set attribute factors that meet the first target value requirement from the attribute factors corresponding to each sequence in the set range and the attribute factors of each sequence in the partial sample data; and determining the video recording subject with defects in the target scene in combination with the video recording subject corresponding to the set attribute factors on the VR video data that needs to be processed.

[0014] In an independently implemented embodiment, the method also includes: determining the attribute factors corresponding to each sequence in the video recording subject of the target scene in combination with the output sample data; determining the video recording subject without defects in the target scene in combination with the video recording subject of the target scene, including: determining the pending attribute factors that meet the second target value requirement among the attribute factors corresponding to each sequence in the video recording subject of the target scene; regarding the video recording subject corresponding to the pending attribute factors as the video recording subject in the middle range; determining the video recording subject without defects in the target scene in combination with the video recording subject in the middle range.

[0015] In an independently implemented embodiment, the method further includes: performing a preliminary screening of the middle range in combination with the video recording subject of the middle range.

[0016] In an independently implemented embodiment, the method also includes: obtaining a configuration image sequence; the configuration image sequence includes a plurality of configuration example data; at least one scene in each of the configuration example data has an abnormality; obtaining a target image sequence; the target image sequence includes a plurality of target example data; each of the target example data includes an abnormal situation in which there is no abnormality in the scene; combining the configuration image sequence and the target image sequence to configure the video classification thread.

[0017] In a second aspect, a VR video playback control system is provided, comprising a processor and a memory communicating with each other, wherein the processor is used to read a computer program from the memory and execute it to implement the above method.

[0018] A VR video playback control method and system provided in an embodiment of the present application determines the video recording subject of a target scene and the video recording subject with defects in the target scene, thereby determining the video recording subject without defects in the target scene based on the video recording subject of the target scene and the video recording subject with defects in the target scene as a distribution in sequence order, so that the distribution without defects can be determined more accurately, so that the confidence of the sample data obtained by performing the optimization operation on the subject without defects is high; in addition, since the video recording subject without defects in the target scene is first determined, and then the optimization operation is performed on the subject without defects, a video optimization result is generated according to the optimization operation, and a video playback instruction is determined in combination with the video optimization result; thereby, accurate and reliable control of video playback can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0020] Figure 1 A flowchart of a VR video playback control method provided in an embodiment of the present application.

[0021] Figure 2 A block diagram of a VR video playback control device provided in an embodiment of the present application.

[0022] Figure 3 This is an architectural diagram of a VR video playback control system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0023] In order to better understand the above technical scheme, the technical scheme of the present application is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present application and the specific features in the embodiments are detailed descriptions of the technical scheme of the present application, rather than limitations on the technical scheme of the present application. In the absence of conflict, the embodiments of the present application and the technical features in the embodiments can be combined with each other.

[0024] See also Figure 1 , shows a VR video playback control method, which may include the technical solutions described in the following steps S101-S103.

[0025] S101. Determine, based on VR video data to be processed, a video recording subject of a target scene and a video recording subject having defects in the target scene.

[0026] The VR video data to be processed may be sample data waiting for optimization operation, and the VR video data to be processed may include local sample data. In the implementation process, the local information in the VR video data to be processed is compared with the VR video data to be processed.

[0027] The video recording subjects of the target scene may include video recording subjects of visible target scenes and video recording subjects of invisible target scenes (i.e., video recording subjects with defects in the target scene).

[0028] The video recording subject of the target scene may represent the video recording subject of the first sequence of the target scene, or may represent the video recording subject of each sequence in the target scene. The video recording subject with defects in the target scene may include the video recording subject of the first sequence with defects in the target scene, or may include the video recording subject of each sequence with defects in the target scene.

[0029] In one possible implementation, determining the video recording subject of the target scene based on VR video data that needs to be processed may include: performing a feature extraction operation on the VR video data that needs to be processed to obtain video event features of the target scene; and determining the video recording subject of the target scene based on the video event features of the target scene.

[0030] In one possible implementation, determining a video recording subject having defects in a target scene based on VR video data that needs to be processed may include: determining a video recording subject having defects in a target scene based on a video classification thread and VR video data that needs to be processed.

[0031] S102: Based on the video recording subjects of the target scene and the video recording subjects with defects in the target scene, determine that there is no video recording subject with defects in the target scene.

[0032] In a possible implementation example, the video recording subjects of the target scene, except for the video recording subjects with defects in the target scene, may be regarded as the video recording subjects without defects in the target scene.

[0033] In one possible implementation, an intermediate range can be determined, and the comparison result between the possibility of each sequence in the intermediate range belonging to the target scene and the possibility of belonging to the non-target scene is lower than the judgment value; among the video recording subjects of the target scene, video recording subjects other than video recording subjects with defects in the target scene and video recording subjects in the intermediate range are regarded as video recording subjects without defects in the target scene.

[0034] The video recording subject having no defects in the target scene may include the video recording subject of the first sequence having no defects in the target scene, or may include the video recording subject of each sequence having no defects in the target scene.

[0035] S103, based on the video recording subjects without defects in the target scene, performing optimization operations on the subjects without defects; generating video optimization results according to the optimization operations, and determining video playback instructions in combination with the video optimization results.

[0036] By recording the subject of the video without the existence of defects in the target scene, the accurate distribution of the absence of defects in the target scene can be obtained, so that the accurate optimization operation without the existence of defects can be performed.

[0037] The target scene range and the defects of the target scene provided in the embodiments of the present application, the target scene range and the defects of the target scene can be applied to the VR video data, video type data, mapped sample data, partial sample data or target sample data that need to be processed mentioned in any embodiment of the present application.

[0038] Taking the target scene as a mouth as an example, the target scene range in the VR video data to be processed can be a combination of range M and range N, and the video recording subject of the target scene can be the video recording subject of each sequence in range M and range N or the video recording subject of each sequence in the first sequence of the combination of range M and range N.

[0039] The defective target scene may be a range N, and the defective video recording subject of the target scene may be a video recording subject of each sequence in the range N or a first video recording subject in the range N.

[0040] In the embodiment of the present application, since the video recording subjects of the target scene and the video recording subjects with defects in the target scene are determined, the video recording subjects without defects in the target scene determined based on the video recording subjects of the target scene and the video recording subjects with defects in the target scene are distributed in a sequence order, so that the distribution without defects can be determined more accurately, so that the confidence of the sample data obtained by performing the optimization operation on the without defects is high; in addition, since the video recording subjects without defects in the target scene are first determined, and then the optimization operation is performed on the without defects, a video optimization result is generated according to the optimization operation, and a video playback instruction is determined in combination with the video optimization result; thereby, accurate and reliable control of video playback can be achieved.

[0041] An implementation process of another VR video playback control method provided in an embodiment of the present application may specifically include the following contents.

[0042] S301: Perform feature extraction on VR video data to be processed to obtain video event features of a target scene.

[0043] S302: Determine a video recording theme of the target scene based on video event features of the target scene.

[0044] S303: Determine the video recording subject with defects in the target scene based on the video classification thread and the VR video data to be processed.

[0045] In a possible implementation, the output sample data can be determined based on the video classification thread and the VR video data to be processed, and the video recording subject with defects in the target scene can be determined based on the output sample data. For example, the target sample data to be input to the video classification thread can be determined based on the VR video data to be processed; then the target sample data is processed by the video classification thread to obtain the output sample data; and the video recording subject with defects in the target scene is determined based on the output sample data.

[0046] In one possible implementation, VR video data to be processed can be loaded into a video classification thread, so that the video classification thread processes the sample data to be processed and obtains output sample data corresponding to the VR video data to be processed, thereby determining the video recording subject with defects in the target scene based on the output sample data corresponding to the VR video data to be processed.

[0047] S304: Based on the video recording subjects of the target scene and the video recording subjects with defects in the target scene, determine that there is no video recording subject with defects in the target scene.

[0048] S305: Based on the video recording subjects without defects in the target scene, an optimization operation is performed on the subjects without defects.

[0049] In the embodiment of the present application, since the video recording subject with defects in the target scene is determined based on the video classification thread, the video recording subject with defects in the target scene can be accurately determined.

[0050] An embodiment of the present application provides another VR video playback control method, which may specifically include the following steps.

[0051] S401: Determine a video recording theme of a target scene based on VR video data to be processed.

[0052] S402: Based on the VR video data to be processed, determine the video category data including the target scene.

[0053] In a possible implementation, based on the VR video data to be processed and the mapping list, mapped sample data can be obtained, and then the mapped sample data can be clustered to obtain video category data.

[0054] In a possible implementation embodiment, the mapping list may be a mapping list of VR video data to be processed to mapped sample data.

[0055] Feature extraction can be performed on the mapped sample data to obtain the video event features of the target scene in the mapped sample data, and then the mapped sample data can be clustered based on the video event features of the target scene in the mapped sample data. In a possible implementation, the mapped sample data can be clustered based on the video event features of the target scene in the VR video data to be processed, or based on the video event features of the target scene in the VR video data to be processed and a mapping list. In an alternative embodiment, the mapped sample data can be clustered based on the distribution information such as the benchmark of the target scene in the VR video data to be processed or the mapped sample data to obtain video category data. The video category data should include the global target scene.

[0056] In a possible implementation, the VR video data to be processed may be processed into clusters to obtain video category data. For example, feature extraction may be performed on the VR video data to be processed to obtain video event features of the target scene in the VR video data to be processed, and based on the video event features of the target scene in the VR video data to be processed, the VR video data to be processed may be processed into clusters to obtain video category data.

[0057] S403: Determine target sample data to be input to the video classification thread based on the video category data.

[0058] In a possible implementation example, video category data may be used as target sample data.

[0059] S404: Process the target sample data using the video classification thread to obtain output sample data.

[0060] S405: Determine the video recording subject with defects in the target scene based on the output sample data.

[0061] S406: Based on the video recording subjects of the target scene and the video recording subjects with defects in the target scene, determine that there is no video recording subject with defects in the target scene.

[0062] S407: Based on the video recording subjects without defects in the target scene, an optimization operation is performed on the subjects without defects.

[0063] Another VR video playback control method provided in an embodiment of the present application may specifically include the following contents.

[0064] S501: Determine a video recording theme of a target scene based on VR video data to be processed.

[0065] S502: Determine the mapped sample data based on the VR video data to be processed and the mapping list.

[0066] S503: Perform clustering processing on the mapped sample data to obtain video category data.

[0067] S504: Determine target sample data to be input to the video classification thread based on the video category data.

[0068] S505: Process the target sample data using the video classification thread to obtain output sample data.

[0069] S506: Determine the video recording subject with defects in the target scene based on the output sample data and the mapping list.

[0070] The output sample data may include: a target list having the same number of sequences as the target sample data, an attribute factor of each attribute in the target list, and a probability of representing the sequence corresponding to each attribute belonging to a target scene or a non-target scene.

[0071] S507 : Based on the video recording subjects of the target scene and the video recording subjects with defects in the target scene, determine that there is no video recording subject with defects in the target scene.

[0072] S508: Based on the video recording subjects without defects in the target scene, an optimization operation is performed on the subjects without defects.

[0073] Another embodiment of the present application provides a VR video playback control method, which may specifically include the following steps.

[0074] S601: Determine a video recording theme of a target scene based on VR video data to be processed.

[0075] S602: Determine the mapped sample data based on the VR video data to be processed and the mapping list.

[0076] S603: Perform clustering processing on the mapped sample data to obtain video category data.

[0077] S604: Determine target sample data to be input to the video classification thread based on the video category data.

[0078] S605: Use the video classification thread to process the target sample data to obtain output sample data.

[0079] S606: Consider the attribute factors of each attribute in the target list as the attribute factors corresponding to each sequence in the video category data.

[0080] S607: Determine the video recording subject with defects in the target scene based on the attribute factors and mapping lists corresponding to each sequence in the video category data.

[0081] During the implementation process, S607 can be achieved in the following ways: based on the attribute factors and mapping lists corresponding to each sequence in the video category data, determine the attribute factors corresponding to each sequence in the partial sample data including the target scene in the VR video data that needs to be processed and corresponding to the video category data; based on the attribute factors corresponding to each sequence in the partial sample data, determine the video recording subject with defects in the target scene.

[0082] Based on the attribute factors and mapping lists corresponding to each sequence in the video category data, it is possible to inversely calculate the attribute factors corresponding to each sequence in the partial sample data corresponding to the video category data in the VR video data that needs to be processed, and then based on the attribute factors corresponding to each sequence in the partial sample data, the video recording subject with defects in the target scene can be determined, thereby providing an easily implemented method for determining the video recording subject with defects in the target scene.

[0083] In some embodiments, a target attribute factor that meets the first target value requirement can be determined on the video recording subject of the target scene from the attribute factors corresponding to each sequence in the partial sample data; a video recording subject corresponding to the first set sequence in the partial sample data and the second set sequence in the VR video data to be processed is determined; and a video recording subject with defects in the target scene is determined based on the video recording subject corresponding to the target attribute factor on the partial sample data and the video recording subject of the second set sequence.

[0084] In one possible implementation, determining a target attribute factor that meets a first target value requirement from attribute factors corresponding to each sequence in partial sample data may include: determining a target attribute factor that is greater than the first target value or less than the second target value from attribute factors corresponding to each sequence in partial sample data.

[0085] The probability that the sequence corresponding to the target attribute factor belongs to the visible target scene is greater than the probability that it belongs to the remaining range outside the visible target scene.

[0086] In an alternative embodiment, the attribute factors corresponding to each sequence in a set range excluding some sample data in the VR video data to be processed can be determined; on the video recording subject of the target scene, the set attribute factors that meet the first target value requirement are determined from the attribute factors corresponding to each sequence in the set range and the attribute factors of each sequence in the partial sample data; based on the video recording subject corresponding to the set attribute factors on the VR video data to be processed, the video recording subject with defects in the target scene is determined.

[0087] In a possible implementation, determining the set attribute factors that meet the first target value requirement from the attribute factors corresponding to each sequence in the set range and the attribute factors of each sequence in the partial sample data may include: determining the set attribute factors that are greater than the first target value or less than the second target value from the attribute factors corresponding to each sequence in the set range and the attribute factors of each sequence in the partial sample data.

[0088] In this way, by determining the attribute factors corresponding to each sequence in a set range excluding some sample data in the VR video data that needs to be processed, the attribute factors corresponding to each sequence of the VR video data that needs to be processed can be obtained, and then based on the size of the attribute factors corresponding to each sequence of the VR video data that needs to be processed, the video recording subject with defects in the target scene in the VR video data that needs to be processed can be determined. This method of determining the video recording subject with defects in the target scene can more quickly screen out abnormal information.

[0089] S608: Based on the video recording subjects of the target scene and the video recording subjects with defects in the target scene, determine that there are no video recording subjects with defects in the target scene.

[0090] S609: Based on the video recording subjects without defects in the target scene, an optimization operation is performed on the subjects without defects.

[0091] In an embodiment of the present application, each attribute factor of the target list output by the video classification thread is regarded as the attribute factor corresponding to each sequence in the video category data, thereby providing a method for determining the attribute factor corresponding to each sequence in the video category data; and based on the attribute factor and mapping list corresponding to each sequence in the video category data, it can be obtained whether each sequence of partial sample data corresponding to the video category data in the VR video data to be processed belongs to the target scene, thereby determining the video recording subject with defects in the target scene through the sequence order, and accurately performing optimization operations on the defective ones.

[0092] Another embodiment of the present application provides a VR video playback control method, which may specifically include the following steps.

[0093] S701: Perform feature extraction on VR video data to be processed to obtain video event features of the target scene.

[0094] S702: Determine a video recording theme of the target scene based on video event features of the target scene.

[0095] S703: Determine video category data including a target scene based on the VR video data to be processed.

[0096] S704: Determine target sample data to be input to the video classification thread based on the video category data.

[0097] S705: Use the video classification thread to process the target sample data to obtain output sample data.

[0098] S706: Determine the video recording subject with defects in the target scene based on the output sample data.

[0099] In one possible implementation, attribute factors corresponding to each sequence in the video recording subject of the target scene can be determined based on the output sample data, and then based on the attribute factors corresponding to each sequence in the video recording subject of the target scene, video recording subjects with defects in the target scene can be determined.

[0100] The attribute factors of each attribute in the target list can be regarded as the attribute factors corresponding to each sequence in the video category data. Then, based on the attribute factors corresponding to each sequence in the video category data and the mapping list, the attribute factors corresponding to each sequence in the partial sample data including the target scene in the VR video data to be processed can be determined. Since the partial sample data is the part of the VR video data to be processed that includes the target scene, the attribute factors corresponding to each sequence in the video recording subject of the target scene can be determined based on the attribute factors corresponding to each sequence in the partial sample data.

[0101] In an alternative embodiment, if the VR video data to be processed is clustered to obtain video category data, the attribute factors corresponding to each sequence in the partial sample data can be determined based on the attribute factors corresponding to each sequence in the video category data, thereby obtaining the attribute factors corresponding to each sequence in the video recording subject of the target scene.

[0102] S707: Determine, among the attribute factors corresponding to the respective sequences in the video recording subject of the target scene, the undetermined attribute factors that meet the second target value requirement.

[0103] Before S707 , a step of determining attribute factors corresponding to each sequence in the video recording subject of the target scene based on the output sample data may be performed.

[0104] S708: Consider the video recording subject corresponding to the undetermined attribute factor as a video recording subject in the middle range.

[0105] S709: Determine that there is no video recording subject with defects in the target scene based on the video recording subjects with defects in the target scene and the video recording subjects in the middle range.

[0106] The target scene may include three ranges, which are respectively: a target scene with defects, an intermediate range, and a target scene without defects. In the video recording subject of the target scene, the video recording subject with defects in the target scene and the video recording subject other than the video recording subject in the intermediate range may be regarded as the video recording subject without defects in the target scene.

[0107] S710: Based on the video recording themes in the middle range, perform a preliminary screening on the middle range.

[0108] In a possible implementation example, step S710 may not be performed.

[0109] In an embodiment of the present application, the video recording subjects in the middle range can be determined first, and the attribute factors corresponding to the respective sequences in the middle range represent that the respective sequences may belong to either the target scene or the non-target scene. Thus, based on the video recording subjects with defects in the target scene and the video recording subjects in the middle range, the video recording subjects without defects in the target scene can be determined, which can reduce the situation where certain sequences in the middle range are regarded as sequences without defects, and thus the video recording subjects without defects in the target scene can be accurately determined.

[0110] An embodiment of the present application provides a video classification thread, and the method may specifically include the following content.

[0111] S801, obtaining a configuration image sequence; the configuration image sequence includes a number of configuration example data; at least one scene in each configuration example data has an abnormality.

[0112] S802, obtaining a target image sequence; the target image sequence includes a plurality of target example data; each target example data includes an abnormal situation in which there is no abnormality in the scene.

[0113] On the basis that at least one scene includes only the target scene, the abnormal situation that there is no abnormality in the scene may be the abnormal situation that there is no abnormality in the target scene. On the basis that at least one scene includes not only the target scene but also the remaining scenes, the abnormal situation that there is no abnormality in the scene includes not only the abnormal situation that there is no abnormality in the target scene but also the abnormal situation that there is no abnormality in the remaining scenes.

[0114] S803: Based on the configuration image sequence and the target image sequence, a video classification thread is configured.

[0115] On the basis that at least one scene includes only the target scene, the obtained video classification thread can determine that each sequence in the input sample data belongs to the visible target scene or the remaining range outside the visible target scene.

[0116] On the basis that at least one scene includes several scenes, the obtained video classification thread can determine that each sequence in the input sample data belongs to each visible scene or the remaining range outside each visible scene. For example, if the VR video data to be processed includes several scenes, the VR video data to be processed can be processed in groups to obtain several video category data, and several target sample data are obtained based on the several video category data, and the several target sample data are loaded into the video classification thread, so as to obtain that each sequence in the several target sample data belongs to each visible scene or the remaining range outside each visible scene, and then obtain that each sequence in the input sample data belongs to each visible scene or the remaining range outside each visible scene.

[0117] In an embodiment of the present application, a plurality of configuration example data are used as configuration inputs, and a plurality of target example data are used as configuration targets, and a video classification thread is configured, so that the subsequently input target sample data can be processed through the video classification thread, and a video recording subject without defects in the target scene is obtained, thereby saving manpower, and the determined video recording subject without defects in the target scene is accurate.

[0118] The method may further include the following steps.

[0119] S1101, obtaining a configuration image sequence; the configuration image sequence includes a number of configuration example data; at least one scene in each configuration example data has an abnormality.

[0120] S1102: Determine the range of the obstruction of each configuration example data.

[0121] S1103: Based on the range of the occluder and the video event features of at least one scene in each configuration example data, determine that at least one scene in each configuration example data has no defects.

[0122] S1104: Determine a target image sequence based on the fact that there are no defects in at least one scene.

[0123] Based on the absence of defects in at least one scene, target example data corresponding to each configuration example data can be determined, the target example data including abnormal situations where there are no abnormalities in the scene and remaining abnormal situations except for the absence of defects in the scene, thereby obtaining a target image sequence.

[0124] S1105 . Based on the configuration image sequence and the target image sequence, a video classification thread is configured.

[0125] Based on the above, please refer to Figure 2, a VR video playback control device 200 is provided, which is applied to a VR video playback control system, and the device comprises:

[0126] A subject determination module 210, for determining a video recording subject of a target scene and a video recording subject having defects in the target scene based on VR video data to be processed;

[0127] The subject recording module 220 is used to combine the video recording subject of the target scene and the video recording subject with defects in the target scene to determine that there is no video recording subject with defects in the target scene;

[0128] The instruction determination module 230 is used to perform optimization operations on the video recording subjects without defects in the target scene; generate video optimization results based on the optimization operations, and determine video playback instructions based on the video optimization results.

[0129] Based on the above, please refer to Figure 3 , shows a VR video playback control system 300, including a processor 310 and a memory 320 that communicate with each other, and the processor 310 is used to read and execute a computer program from the memory 320 to implement the above method.

[0130] Based on the above, a computer-readable storage medium is also provided, on which a computer program stored implements the above method when running.

[0131] In summary, based on the above scheme, since what are determined are the video recording subjects of the target scene and the video recording subjects with defects in the target scene, the video recording subjects without defects in the target scene determined based on the video recording subjects of the target scene and the video recording subjects with defects in the target scene are the distribution in sequence order, so the distribution without defects can be determined more accurately, so that the confidence of the sample data obtained by performing the optimization operation on the without defects is high; in addition, since the video recording subjects without defects in the target scene are determined first, and then the optimization operation is performed on the without defects, a video optimization result is generated according to the optimization operation, and a video playback instruction is determined in combination with the video optimization result; thereby, accurate and reliable control of video playback can be achieved.

[0132] It should be understood that the system and its modules shown above can be implemented in various ways. For example, in some embodiments, the system and its modules can be implemented by hardware, software, or a combination of software and hardware. Among them, the hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or a dedicated design hardware. Those skilled in the art will understand that the above methods and systems can be implemented using computer executable instructions and / or included in a processor control code, such as a carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. Such code is provided on the carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The system and its modules of the present application can not only be implemented by hardware circuits such as ultra-large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., but can also be implemented by software such as executed by various types of processors, and can also be implemented by a combination of the above hardware circuits and software (e.g., firmware).

[0133] It should be noted that different embodiments may produce different beneficial effects. In different embodiments, the beneficial effects that may be produced may be any one or a combination of the above, or any other beneficial effects that may be obtained.

[0134] The basic concepts have been described above. Obviously, for those skilled in the art, the above detailed disclosure is only for example and does not constitute a limitation of the present application. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements and amendments to the present application. Such modifications, improvements and amendments are suggested in the present application, so such modifications, improvements and amendments still belong to the spirit and scope of the exemplary embodiments of the present application.

[0135] At the same time, the present application uses specific words to describe the embodiments of the present application. For example, "one embodiment", "an embodiment", and / or "some embodiments" refer to a certain feature, structure or characteristic related to at least one embodiment of the present application. Therefore, it should be emphasized and noted that "one embodiment" or "an embodiment" or "an alternative embodiment" mentioned twice or more in different positions in this specification does not necessarily refer to the same embodiment. In addition, some features, structures or characteristics in one or more embodiments of the present application can be appropriately combined.

[0136] In addition, it will be appreciated by those skilled in the art that various aspects of the present application may be illustrated and described by a number of patentable categories or situations, including any new and useful process, machine, product or combination of substances, or any new and useful improvements thereto. Accordingly, various aspects of the present application may be performed entirely by hardware, entirely by software (including firmware, resident software, microcode, etc.), or by a combination of hardware and software. The above hardware or software may all be referred to as "data blocks", "modules", "engines", "units", "components" or "systems". In addition, various aspects of the present application may be represented as a computer product located in one or more computer-readable media, which includes computer-readable program code.

[0137] A computer storage medium may include a propagated data signal containing computer program code, for example, in baseband or as part of a carrier wave. The propagated signal may be in a variety of forms, including electromagnetic, optical, etc., or a suitable combination. A computer storage medium may be any computer-readable medium other than a computer-readable storage medium, which can be connected to an instruction execution system, device or apparatus to communicate, propagate or transmit the program for use. The program code on the computer storage medium may be transmitted via any suitable medium, including radio, cable, fiber optic cable, RF, or similar media, or any combination of the above media.

[0138] The computer program codes required for the operation of each part of the present application can be written in any one or more programming languages, including object-oriented programming languages ​​such as Java, Scala, Smalltalk, Eiffel, JADE, Emerald, C++, C#, VB.NET, Python, etc., conventional procedural programming languages ​​such as C language, Visual Basic, Fortran 2003, Perl, COBOL 2002, PHP, ABAP, dynamic programming languages ​​such as Python, Ruby and Groovy, or other programming languages, etc. The program code can be run entirely on the user's computer, or run on the user's computer as an independent software package, or run partially on the user's computer and partially on a remote computer, or run entirely on a remote computer or server. In the latter case, the remote computer can be connected to the user's computer through any network form, such as a local area network (LAN) or a wide area network (WAN), or connected to an external computer (e.g., via the Internet), or in a cloud computing environment, or used as a service such as software as a service (SaaS).

[0139] In addition, unless explicitly stated in the claims, the order of the processing elements and sequences described in this application, the use of alphanumeric characters, or the use of other names are not intended to limit the order of the processes and methods of this application. Although the above disclosure discusses some invention embodiments that are currently considered useful through various examples, it should be understood that such details are only for illustrative purposes, and the attached claims are not limited to the disclosed embodiments. On the contrary, the claims are intended to cover all modifications and equivalent combinations that are consistent with the essence and scope of the embodiments of this application. For example, although the system components described above can be implemented by hardware devices, they can also be implemented only by software solutions, such as installing the described system on an existing server or mobile device.

[0140] Similarly, it should be noted that in order to simplify the description of the disclosure of this application and thus help understand one or more embodiments of the invention, in the above description of the embodiments of this application, multiple features are sometimes combined into one embodiment, figure or description thereof. However, this disclosure method does not mean that the features required by the object of this application are more than the features mentioned in the claims. In fact, the features of the embodiments are less than all the features of the single embodiment disclosed above.

[0141] In some embodiments, numbers describing the number of components and attributes are used. It should be understood that such numbers used for the description of the embodiments are modified by the modifiers "about", "approximately" or "substantially" in some examples. Unless otherwise specified, "about", "approximately" or "substantially" indicate that the numbers allow adaptive changes. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which can be changed according to the required characteristics of individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and adopt the general method of retaining the digits. Although the numerical domains and parameters used to confirm the breadth of the range in some embodiments of the present application are approximate values, in specific embodiments, the setting of such numerical values ​​is as accurate as possible within the feasible range.

[0142] Each patent, patent application, patent application disclosure, and other materials, such as articles, books, instructions, publications, documents, etc., cited in this application are hereby incorporated by reference in their entirety. Except for application history documents that are inconsistent with or conflicting with the content of this application, documents that limit the broadest scope of the claims of this application (currently or later attached to this application) are also excluded. It should be noted that if the descriptions, definitions, and / or use of terms in the attached materials of this application are inconsistent or conflicting with the content described in this application, the descriptions, definitions, and / or use of terms in this application shall prevail.

[0143] Finally, it should be understood that the embodiments described in this application are only used to illustrate the principles of the embodiments of the present application. Other variations may also fall within the scope of the present application. Therefore, as an example and not a limitation, the alternative configurations of the embodiments of the present application may be considered to be consistent with the teachings of the present application. Accordingly, the embodiments of the present application are not limited to the embodiments explicitly introduced and described in the present application.

[0144] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A VR video playback control method, characterized in that: Applied to a video playback control system, the method at least comprises: Determining, based on the VR video data to be processed, a video recording subject of a target scene and a video recording subject having defects in the target scene; Combining the video recording subject of the target scene and the video recording subject with defects in the target scene, determining that there is no video recording subject with defects in the target scene; In combination with the video recording subject without defects in the target scene, an optimization operation is performed on the subject without defects; according to the optimization operation, a video optimization result is generated, and a video playback instruction is determined in combination with the video optimization result; The step of determining a video recording subject of a target scene based on the VR video data to be processed includes: Performing a feature extraction operation on the VR video data to be processed to obtain video event features of the target scene; Determining a video recording theme of the target scene in combination with video event features of the target scene; The determining, based on the VR video data to be processed, the video recording subject having defects in the target scene comprises: determining, based on the video classification thread and the VR video data to be processed, the video recording subject having defects in the target scene; The determining of the video recording subject having defects in the target scene based on the video classification thread and the VR video data to be processed includes: Determine the video type data including the target scene in combination with the VR video data to be processed; In combination with the video category data, determining target sample data to be input to the video classification thread; Processing the target sample data through the video classification thread to obtain output sample data; Determine, in combination with the output sample data, a video recording subject having defects in the target scene; The determining of the video type data including the target scene in combination with the VR video data to be processed includes: Determine the mapped sample data by combining the VR video data to be processed and the mapping list; Performing clustering processing on the mapped sample data to obtain the video category data; The determining the video recording subject having defects in the target scene in combination with the output sample data comprises: determining the video recording subject having defects in the target scene in combination with the output sample data and the mapping list; The output sample data includes: a target list consistent with the number of sequences of the target sample data or the video type data, an attribute factor of each attribute in the target list, and a probability that the sequence corresponding to each attribute belongs to the target scene or the non-target scene; The step of combining the output sample data to determine the video recording subject having defects in the target scene includes: treating the attribute factors of each attribute in the target list as the attribute factors corresponding to each sequence in the video type data; combining the attribute factors corresponding to each sequence in the video type data and the mapping list to determine the video recording subject having defects in the target scene; The combining the attribute factors corresponding to each sequence in the video category data and the mapping list to determine the video recording subject with defects in the target scene includes: Determine, in the VR video data to be processed, attribute factors corresponding to each sequence in the partial sample data including the target scene corresponding to the video category data, by combining the attribute factors corresponding to each sequence in the video category data and the mapping list; Determine the video recording subject with defects in the target scene by combining the attribute factors corresponding to each sequence in the partial sample data; The step of combining the attribute factors corresponding to each sequence in the partial sample data to determine the video recording subject with defects in the target scene includes: On the video recording subject of the target scene, determining a target attribute factor that meets the first target value requirement from the attribute factors corresponding to each sequence in the partial sample data; Determine a first set sequence in the portion of sample data corresponding to a video recording theme of a second set sequence in the VR video data to be processed; Determine the video recording subject with defects in the target scene by combining the video recording subject corresponding to the target attribute factor on the partial sample data and the video recording subject of the second set sequence; The step of combining the attribute factors corresponding to each sequence in the partial sample data to determine the video recording subject with defects in the target scene includes: Determine the attribute factors corresponding to each sequence in a set range of the VR video data to be processed except the part of the sample data; On the video recording subject of the target scene, determine the set attribute factors that meet the first target value requirement from the attribute factors corresponding to each sequence in the set range and the attribute factors of each sequence in the partial sample data; determine the video recording subject with defects in the target scene in combination with the video recording subject corresponding to the set attribute factors on the VR video data to be processed; Wherein, the method further comprises: Determine the attribute factors corresponding to each sequence in the video recording theme of the target scene in combination with the output sample data; The step of combining the video recording subject of the target scene and the video recording subject with defects in the target scene to determine that the video recording subject does not have defects in the target scene includes: determining a pending attribute factor that meets the second target value requirement among the attribute factors corresponding to each sequence in the video recording subject of the target scene; The video recording subject corresponding to the undetermined attribute factor is regarded as a video recording subject in a middle range; Combining the video recording subjects with defects in the target scene and the video recording subjects in the middle range, determining that there are no video recording subjects with defects in the target scene; Wherein, the method further comprises: performing a preliminary screening of the middle range in combination with the video recording subject of the middle range; Wherein, the method further comprises: Obtaining a configuration image sequence; the configuration image sequence includes a plurality of configuration example data; at least one scene in each of the configuration example data has an abnormality; Obtaining a target image sequence; The target image sequence includes a plurality of target example data; each of the target example data includes an abnormal situation in which there is no abnormality in the scene; The video classification thread is configured by combining the configuration image sequence and the target image sequence.

2. A VR video playback control system, characterized in that: The invention comprises a processor and a memory communicating with each other, wherein the processor is used to read a computer program from the memory and execute the computer program to implement the method according to claim 1.

Images