Invention Grant
- Patent Title: Compact neural networks using condensed filters
-
Application No.: US16949994Application Date: 2020-11-23
-
Publication No.: US11763130B2Publication Date: 2023-09-19
- Inventor: Yingzhen Yang , Jianchao Yang , Ning Xu
- Applicant: Snap Inc.
- Applicant Address: US CA Santa Monica
- Assignee: SNAP INC.
- Current Assignee: SNAP INC.
- Current Assignee Address: US CA Santa Monica
- Agency: SCHWEGMAN LUNDBERG & WOESSNER, P.A.
- Main IPC: G06K9/00
- IPC: G06K9/00 ; G06N3/04 ; G06T7/10 ; G06T5/20

Abstract:
A compact neural network system can generate multiple individual filters from a compound filter. Each convolutional layer of a convolutional neural network can include a compound filters used to generate individual filters for that layer. The individual filters overlap in the compound filter and can be extracted using a sampling operation. The extracted individual filters can share weights with nearby filters thereby reducing the overall size of the convolutional neural network.
Public/Granted literature
- US20210073613A1 COMPACT NEURAL NETWORKS USING CONDENSED FILTERS Public/Granted day:2021-03-11
Information query