Search

CN-119697429-B - Video processing method and related device

CN119697429BCN 119697429 BCN119697429 BCN 119697429BCN-119697429-B

Abstract

The embodiment of the specification provides a video processing method and a related device, wherein the video processing method comprises the steps of obtaining an active region video of an object, determining a video acquisition strategy corresponding to the active region video, reading prompt information of the video acquisition strategy, performing video coding on the active region video by a video coder to obtain video characteristics, performing information coding on the prompt information by the information coder to obtain information characteristics, inputting the video characteristics and the information characteristics into a model to perform biological behavior recognition to obtain a recognition result, and generating a tracing video of the object based on the active region video under the condition that the recognition result meets tracing conditions corresponding to the video acquisition strategy. The embodiment of the application can improve the effectiveness of the generated traceability video of the object.

Inventors

  • YANG JIEZHI

Assignees

  • 马上消费金融股份有限公司

Dates

Publication Date
20260512
Application Date
20241122

Claims (10)

  1. 1. A method of video processing, the method comprising: Acquiring an active area video of an object, determining a video acquisition strategy corresponding to the active area video, and reading prompt information of the video acquisition strategy, wherein the video acquisition strategy comprises video acquisition time, prompt information and/or tracing conditions, and the prompt information is used for indicating a model to carry out biological behavior identification on the active area video according to the prompt information; the video encoder performs video encoding on the video of the active area to obtain video characteristics, and the information encoder performs information encoding on the prompt information to obtain information characteristics; Inputting the video features and the information features into the model for biological behavior recognition to obtain a recognition result; and generating a tracing video of the object based on the active region video under the condition that the identification result meets a tracing condition corresponding to the video acquisition strategy, wherein the tracing condition comprises that the identification result is inconsistent with a historical identification result of an associated active region video of the active region video, and the associated active region video comprises the active region video acquired at the previous acquisition time corresponding to the acquisition time of the active region video.
  2. 2. The method of claim 1, wherein the obtaining information features comprises: position coding is carried out on the prompt information to obtain position characteristics, and initial information coding is carried out on the prompt information to obtain initial information characteristics; and carrying out feature fusion on the initial information features and the position features, and inputting fusion features obtained by the feature fusion into an autoregressive model to carry out autoregressive calculation to obtain the information features.
  3. 3. The method according to claim 1, wherein the step of performing biological behavior recognition on the video feature and the information feature input model to obtain a recognition result further comprises, after performing the step of: The related active area video comprises an active area video acquired at the previous acquisition time corresponding to the acquisition time of the active area video; and under the condition that the historical identification result is inconsistent with the identification result, determining that the identification result meets the tracing condition.
  4. 4. The method of claim 3, wherein the generating a traceability video of the object based on the activity area video comprises: reading a first acquisition time of the active area video and a second acquisition time of the associated active area video, and determining a tracing time interval based on the first acquisition time and the second acquisition time; And determining the video of the active area collected in the tracing time interval as the tracing video.
  5. 5. The method of claim 4, wherein the determining that the recognition result satisfies the tracing condition comprises: Reading a time index of the first acquisition time of the video of the active area, and detecting whether the time index is a preset index; If yes, determining that the identification result meets the tracing condition; if not, splitting an associated time interval to obtain at least one sub-time, wherein the associated time interval is determined based on the first acquisition time and the second acquisition time; and generating the traceability video of the object according to the identification result corresponding to the active area video of each sub-time.
  6. 6. The method according to claim 1, wherein the step of performing biological behavior recognition on the video feature and the information feature input model to obtain a recognition result further comprises, after performing the step of: under the condition that the identification result is not null, determining that the identification result meets the tracing condition; The generating the traceability video of the object based on the activity area video comprises the following steps: and determining the active area video as the traceability video.
  7. 7. The method of claim 1, wherein the video encoder video encodes the active area video, and wherein prior to performing the obtaining video feature operation, further comprises: extracting video frames from the video of the active area to obtain a video frame sequence; The sequence of video frames is input to the video encoder.
  8. 8. A video processing apparatus, the apparatus comprising: The system comprises a video acquisition module, a video analysis module and a video analysis module, wherein the video acquisition module is used for acquiring an active area video of an object, determining a video acquisition strategy corresponding to the active area video and reading prompt information of the video acquisition strategy, and the video acquisition strategy comprises video acquisition time, video acquisition duration, prompt information and/or tracing conditions, and the prompt information is used for indicating a model to carry out biological behavior identification on the active area video according to the prompt information; The coding module is used for video coding the video of the active area by the video coder to obtain video characteristics, and information coding the prompt information by the information coder to obtain information characteristics; The biological behavior recognition module is used for inputting the video features and the information features into the model to perform biological behavior recognition, so as to obtain a recognition result; And under the condition that the identification result meets the tracing condition corresponding to the video acquisition strategy, operating a video generation module, wherein the video generation module is used for generating a tracing video of the object based on the active region video, the tracing condition comprises that the identification result is inconsistent with the historical identification result of the associated active region video of the active region video, and the associated active region video comprises the active region video acquired in the previous acquisition time corresponding to the acquisition time of the active region video.
  9. 9. A video processing apparatus, the apparatus comprising: A processor, and a memory configured to store computer-executable instructions that, when executed, cause the processor to perform the video processing method of any of claims 1-7.
  10. 10. A computer readable storage medium for storing computer executable instructions which, when executed by a processor, implement the video processing method of any of claims 1-7.

Description

Video processing method and related device Technical Field The present application relates to the field of data processing technologies, and in particular, to a video processing method and a related device. Background With the continuous development and progress of the internet and information technology, more and more services are performed online, and with the increasing of living standard, the requirements of users on articles or foods used in daily life are higher and higher, based on the requirements, in order to improve the perception degree of the users on the articles or foods used, tracing becomes an important way for a provider or a recommendation direction user to recommend the articles or foods, and how to improve the tracing convenience and efficiency is an important point of increasing attention of the provider, the recommendation direction and the user. Disclosure of Invention In a first aspect, an embodiment of the present application provides a video processing method, including: Acquiring an active area video of an object, determining a video acquisition strategy corresponding to the active area video, and reading prompt information of the video acquisition strategy; the video encoder performs video encoding on the video of the active area to obtain video characteristics, and the information encoder performs information encoding on the prompt information to obtain information characteristics; Inputting the video features and the information features into a model for biological behavior recognition to obtain recognition results; And generating a traceability video of the object based on the active area video under the condition that the identification result meets the traceability condition corresponding to the video acquisition strategy. It can be seen that, in the embodiment of the present application, after an active area video of an object is obtained in the process of performing video processing, a video acquisition strategy corresponding to the active area video is first determined, and then prompt information of the video acquisition strategy is read, so that the active area video corresponds to the video acquisition strategy, subsequent biological behavior recognition is performed through prompt information prompt of the video acquisition strategy, flexibility and effectiveness of biological behavior recognition performed on the active area video are improved, on this basis, a video encoder performs video encoding on the active area video to obtain video features, an information encoder performs information encoding on the prompt information to obtain information features, then the video features and the information features are input into a model to perform biological behavior recognition, recognition results are obtained, and therefore, before the biological behavior recognition is performed through the model, the active area video and the prompt information are respectively encoded, accurate encoding results of the active area video and the prompt information are obtained, then the video features and the information features obtained through encoding are used as input to enable the acquisition of the model to perform the generation of behavior recognition, accuracy of the biological behavior recognition of the model is improved, on the basis of the video corresponding to perform the biological behavior recognition by the model, the video acquisition strategy is obtained, the information recognition results are obtained by the video is traced back to the basis of the biological behavior recognition strategy, the video is generated, the condition is generated, and the video is traced back to the condition is generated. In a second aspect, an embodiment of the present application provides a video processing apparatus, including: the video acquisition module is used for acquiring an active area video of an object, determining a video acquisition strategy corresponding to the active area video, and reading prompt information of the video acquisition strategy; The coding module is used for video coding the video of the active area by the video coder to obtain video characteristics, and information coding the prompt information by the information coder to obtain information characteristics; The biological behavior recognition module is used for inputting the video features and the information features into the model to perform biological behavior recognition, so as to obtain a recognition result; And under the condition that the identification result meets the tracing condition corresponding to the video acquisition strategy, operating a video generation module, wherein the video generation module is used for generating the tracing video of the object based on the active area video. In a third aspect, an embodiment of the present application provides a video processing apparatus comprising a processor, and a memory configured to store computer-executable instructions that, when executed