CVPR2023

SparseViT: Revisiting Activation Sparsity for Efficient High-Resolution Vision Transformer
High-resolution images enable neural networks to learn richer visual representations. However, this improved performance comes at the …
PDF Cite Code Project Poster
FlatFormer: Flattened Window Attention for Efficient Point Cloud Transformer
Transformer, as an alternative to CNN, has been proven effective in many modalities (e.g., texts and images). For 3D point cloud …
PDF Cite Code Project Poster