Video Anonymization via Key Frame Extraction and Object Region Replacement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video anonymization technologies face challenges with high computing power requirements and low efficiency due to large resource consumption, making it difficult to integrate anonymization algorithms in vehicle chip hardware resources.
Innovation Solution
A video anonymization method that extracts key video frames, performs key object recognition, and replaces detected key image regions with predetermined images, reducing the workload and computing power demands by partially anonymizing videos without full processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If existing anonymization algorithms are integrated in cloud computing centers, then anonymization can be performed, but computing power resources are large and efficiency is low
Solution Approach 1:
The patent segments the video processing task by extracting only key video frames at specific intervals rather than processing every frame. This segmentation approach reduces the total number of frames requiring anonymization while maintaining privacy protection effectiveness, thereby improving productivity while reducing computing power consumption.
Solution Approach 2:
The patent applies partial action by selectively anonymizing only key frames that contain important visual information, rather than performing full anonymization on all video frames. This partial processing approach maintains adequate privacy protection while significantly reducing the computing power resources required and improving overall anonymization efficiency.
2Ease of operation
If anonymization algorithms are integrated in vehicle chip hardware resources, then processing can be done locally, but hardware resources are limited due to multifunctional integration
Solution Approach 1:
By segmenting video processing into key frame extraction and anonymization steps, the patent reduces the computational load to a level that can be handled by limited vehicle chip hardware resources, enabling local processing without requiring excessive hardware capacity.
Solution Approach 2:
The patent implements partial processing by anonymizing only key frames rather than all frames, which reduces the quantity of computational work to match the limited hardware resources available in vehicle chips, while still providing effective local privacy protection.
3Reliability
If all video frames are processed for anonymization, then complete privacy protection is achieved, but workload and computing power requirements are high
Solution Approach 1:
The patent segments the video stream into key frames at specific intervals for anonymization processing. This segmentation maintains adequate privacy protection reliability by capturing essential visual information while significantly reducing processing workload and device complexity compared to full-frame processing.
Solution Approach 2:
The patent applies partial action by processing only key frames rather than all video frames. This approach achieves sufficient privacy protection reliability for practical applications while dramatically reducing the processing workload and simplifying device requirements.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A video anonymization method and apparatus, an electronic device, and a storage medium are disclosed. The method includes: extracting a plurality of first video frames from a to-be-processed video to obtain a first video frame sequence; performing key object recognition on the first video frames; determining, if key objects are recognized from the first video frames, an image region where the key objects are located as a detected key image region, and replacing original images therein with predetermined images to obtain first anonymized frames; and replacing the first video frames in the first video frame sequence until all the first video frames have been replaced to obtain a first anonymized frame sequence. According to the method, video frames in a to-be-processed video are partially anonymized to anonymize the video without being processed wholly, thus reducing the workload of anonymization, reducing excessive demands for platform computing power, effectively solving the technical problem of large computing power resources required by the existing anonymization, and effectively improving the efficiency of anonymization as the number of to-be-processed frames is reduced.