Vision gnn: An image is worth graph of nodes

K Han, Y Wang, J Guo, Y Tang… - Advances in neural …, 2022 - proceedings.neurips.cc
Network architecture plays a key role in the deep learning-based computer vision system.
The widely-used convolutional neural network and transformer treat the image as a grid or …
被引用次数：257 相关文章

Vision GNN: An Image is Worth Graph of Nodes

K Han, Y Wang, J Guo, Y Tang, E Wu - Advances in Neural …, 2022 - openreview.net
Network architecture plays a key role in the deep learning-based computer vision system.
The widely-used convolutional neural network and transformer treat the image as a grid or …

Vision GNN: An Image is Worth Graph of Nodes

K Han, Y Wang, J Guo, Y Tang, E Wu - proceedings.neurips.cc
In our ViG block, we propose to increase feature diversity in nodes by utilizing more feature
transformation such as FFN module. We show the empirical comparison between vanilla …

Vision GNN: An Image is Worth Graph of Nodes

K Han, Y Wang, J Guo, Y Tang, E Wu - arXiv e-prints, 2022 - ui.adsabs.harvard.edu
Network architecture plays a key role in the deep learning-based computer vision system.
The widely-used convolutional neural network and transformer treat the image as a grid or …

Vision GNN: An Image is Worth Graph of Nodes

K Han, Y Wang, J Guo, Y Tang, E Wu - arXiv preprint arXiv:2206.00272, 2022 - arxiv.org
Network architecture plays a key role in the deep learning-based computer vision system.
The widely-used convolutional neural network and transformer treat the image as a grid or …
[PDF] neurips.cc

[PDF][PDF] Vision GNN: An Image is Worth Graph of Nodes

K Han, Y Wang, J Guo, Y Tang, E Wu - papers.neurips.cc
Network architecture plays a key role in the deep learning-based computer vision system.
The widely-used convolutional neural network and transformer treat the image as a grid or …

Vision GNN: an image is worth graph of nodes

K Han, Y Wang, J Guo, Y Tang, E Wu - Proceedings of the 36th …, 2022 - dl.acm.org
Network architecture plays a key role in the deep learning-based computer vision system.
The widely-used convolutional neural network and transformer treat the image as a grid or …

Vision GNN: An Image is Worth Graph of Nodes

K Han, Y Wang, J Guo, Y Tang, E Wu - papers.neurips.cc
In our ViG block, we propose to increase feature diversity in nodes by utilizing more feature
transformation such as FFN module. We show the empirical comparison between vanilla …