Pedestrian attribute recognition system and method based on multi-layer feature learning
A technology of attribute recognition and feature learning, applied in the field of computer vision, can solve the problems of not being able to learn, not paying attention to learning, not being able to better use and integrate multi-layer features, etc., to achieve the effect of enhancing experimental results and predicting results
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment
[0046] Such as figure 1 , figure 2 , image 3 As shown, the present embodiment provides a pedestrian attribute recognition system based on multi-layer feature learning, including a pedestrian attribute recognition network training module and a pedestrian attribute recognition system test module, and the pedestrian attribute recognition network training module includes bottom-up feature extraction module, top-down feature fusion module, feature prediction module and multi-layer prediction fusion module.
[0047]In this embodiment, the bottom-up feature extraction module: this module is based on a classic convolutional neural network, as a module that processes pictures layer by layer to obtain multi-layer features. This embodiment selects the classic network ResNet- 50 as the main body of the module, and the four Residual Blocks (residual modules) in ResNet-50 are used as the layer of feature output. The structure of the residual module is as follows Figure 4 As shown, t...
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