As part of our Lecture Series in Intelligent Perception and Computing, Prof. Jingdong Wang was invited to give a talk on Friday, 2nd July 2021 at CRIPAC. Prof. Wang is a Senior Principal Research Manager with the Visual Computing Group at Microsoft Research Asia.

Prof. Wang’s talk was entitled “Local Transformer Attention is Equivalent to Convolution”. Local Vision Transformer has been attracting a lot of interest. The major component in Local Vision Transformer, local attention, performs the attention separately over small local windows. Prof. Wang presented the analysis of local attention from sparse connectivity, weight sharing and dynamic weight. He showed that local attention is equivalent to inhomogeneous dynamic depth-wise convolution. Experimental results on ImageNet classification, COCO object detection, and ADE segmentation indicated that inhomogeneous dynamic depth-wise convolution using convolution for dynamic weight prediction outperformed local vision transformer, Swin Transformer, in the tiny model case, and performed almost the same for the base model case. In addition, he provided a relation graph to explain the relations between various networks, including recently-developed MLP-based models, ViT, and convolution-based models.

This talk was very popular and all seats were occupied. Many attendees raised their own questions and had in-depth discussions with Prof.Wang.

Biography of Prof. Jingdong Wang:

Jingdong Wang is a Senior Principal Research Manager with the Visual Computing Group at Microsoft Research Asia, Beijing, China. He received the B.Eng. and M.Eng. degrees from the Department of Automation at Tsinghua University in 2001 and 2004, respectively, and the PhD degree from the Department of Computer Science and Engineering, the Hong Kong University of Science and Technology, Hong Kong, in 2007. His areas of interest include neural network design, human pose estimation, large-scale indexing, and person re-identification. He is/was an Associate Editor of the IEEE TPAMI, the IEEE TMM and the IEEE TCSVT, and is an area chair of several leading Computer Vision and AI conferences, such as CVPR, ICCV, ECCV, ACM MM, IJCAI, and AAAI. He was elected as an IAPR Fellow, an ACM Distinguished Member, and an Industrial Distinguished Lecturer Program (iDLP) speaker of the IEEE Circuits and Systems Society.