Timezone: »

Green Hierarchical Vision Transformer for Masked Image Modeling
Lang Huang · Shan You · Mingkai Zheng · Fei Wang · Chen Qian · Toshihiko Yamasaki

Wed Nov 30 09:00 AM -- 11:00 AM (PST) @ Hall J #620
We present an efficient approach for Masked Image Modeling (MIM) with hierarchical Vision Transformers (ViTs), allowing the hierarchical ViTs to discard masked patches and operate only on the visible ones. Our approach consists of three key designs. First, for window attention, we propose a Group Window Attention scheme following the Divide-and-Conquer strategy. To mitigate the quadratic complexity of the self-attention w.r.t. the number of patches, group attention encourages a uniform partition that visible patches within each local window of arbitrary size can be grouped with equal size, where masked self-attention is then performed within each group. Second, we further improve the grouping strategy via the Dynamic Programming algorithm to minimize the overall computation cost of the attention on the grouped patches. Third, as for the convolution layers, we convert them to the Sparse Convolution that works seamlessly with the sparse data, i.e., the visible patches in MIM. As a result, MIM can now work on most, if not all, hierarchical ViTs in a green and efficient way. For example, we can train the hierarchical ViTs, e.g., Swin Transformer and Twins Transformer, about 2.7$\times$ faster and reduce the GPU memory usage by 70%, while still enjoying competitive performance on ImageNet classification and the superiority on downstream COCO object detection benchmarks.

Author Information

Lang Huang (The University of Tokyo)

I am currently a second-year Ph.D. student at the Department of Information & Communication Engineering, The University of Tokyo. Prior to that, I received a Master’s degree from the Department of Machine Intelligence, School of Electronics Engineering and Computer Science, Peking University in 2021. My research interests include self-supervised representation learning, robust learning from noisy data, and vision transformers.

Shan You (SenseTime Research)
Mingkai Zheng (University of Sydney)
Fei Wang (Sensetime)
Chen Qian (SenseTime)
Toshihiko Yamasaki (The University of Tokyo)

More from the Same Authors