semantic_segmentation

Applying TransXNet to Semantic Segmentation

For details, please address "TransXNet: Learning Both Global and Local Dynamics with a Dual Dynamic Token Mixer for Visual Recognition".

1. Requirements

We highly suggest using our provided dependencies to ensure reproducibility:

# Environments:
cuda==11.3
python==3.8.15
# Packages:
mmcv==1.7.1
timm==0.6.12
torch==1.12.1
torchvision==0.13.1
mmsegmentation==0.30.0

2. Data Preparation

Prepare ADE20K according to the guidelines.

3. Main Results on ADE20K with Pretrained Models

Method	Backbone	Pretrain	Iters	mIoU	Config	Download
Semantic FPN	TransXNet-T	ImageNet-1K	80K	45.5	config	log & model
Semantic FPN	TransXNet-S	ImageNet-1K	80K	48.5	config	log & model
Semantic FPN	TransXNet-B	ImageNet-1K	80k	49.9	config	log & model

4. Train

To train TransXNet + Semantic FPN models on ADE20K with 8 gpus (single node), run:

bash scripts/train_sfpn_transxnet_tiny.sh # train TransXNet-T + Semantic FPN
bash scripts/train_sfpn_transxnet_small.sh # train TransXNet-S + Semantic FPN
bash scripts/train_sfpn_transxnet_base.sh # train TransXNet-B + Semantic FPN

5. Validation

To evaluate TransXNet + Semantic FPN models on ADE20K, run:

# Take TransXNet-T + Semantic FPN as an example:
python3 -m torch.distributed.launch \
--nproc_per_node=8 \
--master_port=$((RANDOM+8888)) \
test.py \
configs/sfpn_transxnet_tiny.py \
path/to/checkpoint \
--out work_dirs/output.pkl \
--eval mIoU \
--launcher pytorch

Citation

If you find this project useful for your research, please consider citing:

@article{lou2023transxnet,
  title={TransXNet: Learning Both Global and Local Dynamics with a Dual Dynamic Token Mixer for Visual Recognition},
  author={Meng Lou and Shu Zhang and Hong-Yu Zhou and Sibei Yang and Chuan Wu and Yizhou Yu},
  journal={IEEE Transactions on Neural Networks and Learning Systems},
  year={2025}
}

Contact

If you have any questions, please feel free to create issues or contact me at lmzmm.0921@gmail.com.