Search papers, labs, and topics across Lattice.
College of Computer Science, Beijing University of Technology, Beijing, 100124, China
1
0
0
3
DMFormer is introduced, an innovative multi-modal BEV perception framework that employs Transformer architecture and a diffusion denoising model to tackle key challenges, including sensor noise, efficient fusion of multi-modal data, and modeling dynamic scenes.
