Four-value weight and multiple classification-based human face feature extraction method
A multi-classification and face feature technology, applied in the field of image processing, can solve the problems of reducing computing speed, increasing parameters, slowing down speed, etc., to achieve the effect of solving insufficient storage space, solving huge consumption, and improving accuracy
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment Construction
[0022] In order to clearly illustrate the technical features of the solution, the solution will be described below through specific implementation modes.
[0023] see figure 1 , the embodiment of the present invention provides a method for extracting facial features based on four-valued weights and multiple classifications, including:
[0024] Step S1: collect face image samples to build a face training sample database, and all face image samples of the same person are used as a group;
[0025] Step S2: building a convolutional neural network with multiple classification functions;
[0026] Step S3: Adjust the caffe framework to convert the floating-point weights of the convolutional neural network into four-value weights;
[0027] Step S4: Preprocess the face image samples in the face training sample database to generate a mean value file, and put the face image samples into the convolutional neural network with parameters set, and perform the training under the adjusted ca...
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