Visual Attention Network (VAN) for Segmentaion

This repo is a PyTorch implementation of applying VAN (Visual Attention Network) to semantic segmentation. The code is based on mmsegmentaion.

More detailes can be found in Visual Attention Network.

Citation:

@article{guo2022visual,
  title={Visual Attention Network},
  author={Guo, Meng-Hao and Lu, Cheng-Ze and Liu, Zheng-Ning and Cheng, Ming-Ming and Hu, Shi-Min},
  journal={arXiv preprint arXiv:2202.09741},
  year={2022}
}

Results

Notes: Pre-trained models can be found in Visual Attention Network for Classification.

VAN + UperNet

Backbone	Iters	mIoU	Config	Download
VAN-Tiny	160K	41.1	config
VAN-Small	160K	44.9	config
VAN-Base	160K	48.3	config
VAN-Large	160K	50.1	config

Notes: In this scheme, we use multi-scale validation following Swin-Transformer.

VAN + Semantic FPN

Backbone	Iters	mIoU	Config	Download
VAN-Tiny	40K	38.5	config	Google Drive
VAN-Small	40K	42.9	config	Google Drive
VAN-Base	40K	46.7	config	Google Drive
VAN-Large	40K	48.1	config	Google Drive

Preparation

Install MMSegmentation and download ADE20K according to the guidelines in MMSegmentation.

Requirement

Pytorch >= 1.7
MMSegmentation == v0.12.0 (https://github.com/open-mmlab/mmsegmentation/tree/v0.12.0)

Training

We use 8 GPUs for training by default. Run:

dist_train.sh /path/to/config 8

Evaluation

To evaluate the model, run:

dist_test.sh /path/to/config /path/to/checkpoint_file 8 --out results.pkl --eval mIoU

Acknowledgment

Our implementation is mainly based on mmsegmentaion, Swin-Transformer, and PoolFormer. Thanks for their authors.

LICENSE

This repo is under the Apache-2.0 license. For commercial use, please contact the authors.