Skip to content

Latest commit

 

History

History
132 lines (98 loc) · 5.45 KB

README.md

File metadata and controls

132 lines (98 loc) · 5.45 KB
Raw

Deep Multimodal Fusion by Channel Exchanging

By Yikai Wang, Wenbing Huang, Fuchun Sun, Tingyang Xu, Yu Rong, Junzhou Huang.

[Paper] [Paper & Appendix] (with proofs and visualizations)

[Slides] [Poster] [BibTex]

This repository is an official PyTorch implementation of "Deep Multimodal Fusion by Channel Exchanging", in NeurIPS 2020. Its extension for multimodal and multitask learning has been accepted by TPAMI 2023 (arXiv).

The basic method and applications are introduced as follows:

If you find our work useful for your research, please consider citing the following papers.

@inproceedings{wang2020cen,
  title={Deep Multimodal Fusion by Channel Exchanging},
  author={Wang, Yikai and Huang, Wenbing and Sun, Fuchun and Xu, Tingyang and Rong, Yu and Huang, Junzhou},
  booktitle={Advances in Neural Information Processing Systems (NeurIPS)},
  year={2020}
}

@article{wang2023cenpami,
  title={Channel Exchanging Networks for Multimodal and Multitask Dense Image Prediction},
  author={Wang, Yikai and Sun, Fuchun and Huang, Wenbing and He, Fengxiang and Tao, Dacheng},
  journal={IEEE Transaction on Pattern Analysis and Machine Intelligence (TPAMI)},
  year={2023}
}

Datasets

For semantic segmentation task on NYUDv2 (official dataset), we provide a link to download the dataset here. The provided dataset is originally preprocessed in this repository, and we add depth data in it.

For image-to-image translation task, we use the sample dataset of Taskonomy, where a link to download the sample dataset is here.

Please modify the data paths in the codes, where we add comments 'Modify data path'.

Dependencies

python==3.6.2
pytorch==1.0.0
torchvision==0.2.2
imageio==2.4.1
numpy==1.16.2
scikit-learn==0.20.2
scipy==1.1.0
opencv-python==4.0.0

Semantic Segmentation

First,

cd semantic_segmentation

Training script for segmentation with RGB and Depth input, the default setting uses RefineNet (ResNet101),

python main.py --gpu 0 -c exp_name  # or --gpu 0 1 2

Or, for training RefineNet with ResNet152,

python main.py --gpu 0 --enc 152 --num-epoch 150 150 150 -c exp_name  # or --gpu 0 1 2

Evaluation script,

python main.py --gpu 0 --resume path_to_pth --evaluate  # optionally use --save-img to visualize results

Checkpoint models, training logs and the single-scale performance on NYUDv2 (with RefineNet) are provided as follows. We also provide an additional comparison with our ViT-based TokenFusion in CVPR 2022:

Method Backbone Pixel Acc. (%) Mean Acc. (%) Mean IoU (%) Download
CEN ResNet101 76.2 62.8 51.1 Google Drive
CEN ResNet152 77.0 64.4 51.6 Google Drive
TokenFusion SegFormer-B3 78.7 67.5 54.8 Google Drive

Image-to-Image Translation

First,

cd image2image_translation

Training script, an example of translation from Shade (2) and Texture (7) to RGB (0) (could reach 62~63 FID score),

python main.py --gpu 0 --img-types 2 7 0 -c exp_name

This script will auto-evaluate on the validation dataset every 5 training epochs.

Predicted images will be automatically saved during training, in the following folder structure:

code_root/ckpt/exp_name/results
  ├── input0  # 1st modality input
  ├── input1  # 2nd modality input
  ├── fake0   # 1st branch output 
  ├── fake1   # 2nd branch output
  ├── fake2   # ensemble output
  ├── best    # current best output
  │    ├── fake0
  │    ├── fake1
  │    └── fake2
  └── real    # ground truth output

For training with other modalities, the index for each img-type is described as follows, and also in Line 69 of main.py.

0: 'rgb', 1: 'normal', 2: 'reshading', 3: 'depth_euclidean', 
4: 'depth_zbuffer', 5: 'principal_curvature', 6: 'edge_occlusion', 
7: 'edge_texture', 8: 'segment_unsup2d', 9: 'segment_unsup25d'

Full quantitative results are provided in the paper.

Compared with ViT-based TokenFusion in CVPR 2022:

Method Task FID KID Download
CEN Texture+Shade->RGB 62.6 1.65 -
TokenFusion Texture+Shade->RGB 45.5 1.00 Google Drive

License

CEN is released under MIT License.