Data generation method and device, electronic equipment and storage medium
By combining a multi-sensor temporal diffusion model and an autoencoder, the problem of insufficient cross-modal consistency in multimodal sensor data generation is solved, and high-quality data generation is achieved.
CN122457748APending Publication Date: 2026-07-24BEIJING CO WHEELS TECH CO LTD
View PDF 0 Cites 0 Cited by
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- BEIJING CO WHEELS TECH CO LTD
- Filing Date
- 2026-02-12
- Publication Date
- 2026-07-24
Smart Images
Figure CN122457748A_ABST
Abstract
The application discloses a data generation method and device, electronic equipment and storage medium. The data generation method comprises the following steps: utilizing a multi-sensor time sequence diffusion model to perform joint denoising processing on target point cloud noise and target video noise, so as to obtain target point cloud potential features and target video potential features, wherein the joint denoising processing comprises inter-sensor attention operation and intra-sensor attention operation; utilizing a first decoder in a point cloud variational autoencoder to decode and render the target point cloud potential features, so as to obtain target point cloud data; and utilizing a second decoder in a video variational autoencoder to decode the target video potential features, so as to obtain target video data. The application improves the overall efficiency and data quality of data generation.
Need to check novelty before this filing date? Find Prior Art