Compression method and system based on local feature descriptor of video

Abstract

The invention discloses a compression method and system based on a local feature descriptor of a video and is used for solving a compression problem of a local feature descriptor of a video in the prior art. The method comprises prediction, quantization and entropy coding, wherein the prediction comprises intra-frame prediction, inter-frame prediction and reconstructed frame prediction, an optimal prediction signal is selected from prediction signals of three modes through a preset optimization strategy, a residual coefficient of the prediction signals is calculated, the quantization is performed on the residual coefficient to obtain a quantization coefficient, the entropy coding is performed on the quantization coefficient, a code stream formed after the entropy coding is output, and compression of the local feature descriptor of the video is finished. The local feature descriptor of the video is compressed through the compression method and system, compressed local feature data of the video is compactly expressed on the basis of the combined video content, high compression ratio is achieved, and the retrieval efficiency of the video data is improved.

Description

technical field [0001] The invention belongs to the technical field of data processing, and in particular relates to a video-based local feature descriptor compression method and system. Background technique [0002] The local feature starts from the local structure of the image / video, describes the regional information in the image / video, and embodies the unique descriptiveness, thus transforming the complicated image / video matching problem into the measurement problem of the feature vector. Therefore, the local feature descriptor It has been widely used in image / video data processing and computer vision, such as visual retrieval, augmented reality, and target detection and recognition. [0003] In the prior art, common local feature descriptors include SIFT (Scale-Invariant Feature Transform), SURF (Speeded Up Robust Features), and GLOH (Gradient Location and Orientation Histogram), and the huge data volume of the original local feature descriptor is Describe one of the m...

AI Technical Summary

Problems solved by technology

However, the research on video-based local feature descriptor compression methods is still in its infancy, which has caused great obstacles to the development of video data retrieval technology.

Images