Target identification method, device, apparatus and storage medium
By pre-setting convolution kernel parameters to determine the output features of 3D point clouds and inversely deriving the effective output points and their input sequence, the problem of excessive memory consumption of sparse convolution on the NPU is solved, the target recognition efficiency is improved, and it is suitable for autonomous driving and medical image processing.
CN122244572APending Publication Date: 2026-06-19YUAN LI (BEI JING) BAN DAO TI JI SHU YOU XIAN GONG SI
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- YUAN LI (BEI JING) BAN DAO TI JI SHU YOU XIAN GONG SI
- Filing Date
- 2026-02-09
- Publication Date
- 2026-06-19
Smart Images
Figure CN122244572A_ABST
Abstract
This application discloses a target recognition method, apparatus, device, and storage medium. The method includes: acquiring three-dimensional point cloud data; determining three-dimensional point cloud output features corresponding to the three-dimensional point cloud data based on preset convolution kernel parameters; determining valid output points and corresponding valid input sequence sequences of the three-dimensional point cloud data based on the three-dimensional point cloud output features; determining valid output features corresponding to the three-dimensional point cloud data based on the valid output points and corresponding valid input sequence sequences; and recognizing a target to be recognized based on the valid output features to obtain a target recognition result. This can reduce the memory usage of three-dimensional point cloud data during computation, thereby improving target recognition efficiency.
Need to check novelty before this filing date? Find Prior Art