A Pedestrian Feature Extraction and Representation Method Based on Area Sparse Integral Channel
A technology of integrating channel features and integrating channels, which is applied in the field of feature extraction and representation, can solve the problems of strong discriminative power, weak discriminative power, and poor real-time performance of mixed features, and achieves the effect of strong discriminative ability, few setting parameters, and fast calculation speed.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment Construction
[0028] The present invention provides a pedestrian feature extraction and representation method based on the regional sparse integration channel, which firstly utilizes the spatial layout information of uniformly randomly sampled regions and subregions, secondly utilizes various heterogeneous source information in each subregion, and then uses the max pooling algorithm The sub-area features are fused into area features, and finally all the area features are concatenated to be pedestrian descriptors. The pedestrian features extracted by this method have strong discriminative ability, fast calculation speed, less setting parameters, and take into account the advantages of rigid targets and deformable targets, which are suitable for fast pedestrian detection or pedestrian tracking.
[0029] In order to describe the content of the present invention conveniently, some terms are defined first.
[0030] Definition 1: Integral graph. For a grayscale image, the value of any point (x, ...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com