Binary neural network-based local activation method and system
US12400441B2 · kind B2 · utility
Assignee
Inventors
Key dates
| Filing date | May 18, 2021 |
| Grant date | Aug 26, 2025 |
| Priority date | — |
| Expiry date | May 17, 2042 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06V10/806
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
A local activation method and system based on a binary neural network, said method including: during forward propagation, comparing the difference between a center pixel and an adjacent pixel to determine a local activation value; during forward propagation, setting an appropriate number of local activation channels and an activation direction to obtain a local activation feature map having different activation directions; during forward propagation, using weighting coefficients, which can be learned, to perform channel fusion on the output feature map after local activation and direct activation, and obtaining an output feature map containing both texture features and contour features; during backpropagation, using an asymptotic sine function to update the weights of the binary neural network. The invention is capable of effectively reducing the loss of information during binary activation, and can effectively reduce gradient mismatch during backward gradient update of a binary neural network, thereby improving the performance of the binary neural network.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.