DIS/README.md

185 lines
7.0 KiB
Markdown
Raw Normal View History

2022-03-07 08:20:41 +01:00
<p align="center">
2022-03-07 08:25:14 +01:00
<img width="420" height="320" src="figures/dis-logo-official.png">
2022-03-07 08:20:41 +01:00
</p>
2022-03-08 09:16:49 +01:00
![dis5k-v1-sailship](figures/dis5k-v1-sailship.jpeg)
<br>
2022-07-08 07:26:22 +02:00
## [Highly Accurate Dichotomous Image Segmentation ECCV 2022](https://arxiv.org/pdf/2203.03041.pdf)
[Xuebin Qin](https://xuebinqin.github.io/), [Hang Dai](https://scholar.google.co.uk/citations?user=6yvjpQQAAAAJ&hl=en), [Xiaobin Hu](https://scholar.google.de/citations?user=3lMuodUAAAAJ&hl=en), [Deng-Ping Fan*](https://dengpingfan.github.io/), [Ling Shao](https://scholar.google.com/citations?user=z84rLjoAAAAJ&hl=en), [Luc Van Gool](https://scholar.google.com/citations?user=TwMib_QAAAAJ&hl=en).
<br>
2022-03-08 09:16:49 +01:00
## This is the official repo for our newly formulated DIS task:
[**Project Page**](https://xuebinqin.github.io/dis/index.html), [**Arxiv**](https://arxiv.org/pdf/2203.03041.pdf).
2022-03-07 08:20:41 +01:00
<br>
2022-03-08 09:11:23 +01:00
2022-07-05 20:33:10 +02:00
## Updates !!!
<br>
## ** (2022-Jul.-17)** Our paper, code and dataset are now officially released!!! Please check our project page for more details: [**Project Page**](https://xuebinqin.github.io/dis/index.html).<br>
2022-07-05 20:41:21 +02:00
** (2022-Jul.-5)** Our DIS work is now accepted by ECCV 2022, the code and dataset will be released before July 17th, 2022. Please be aware of our updates.
2022-07-05 20:33:10 +02:00
<br>
## 1. [Our DIS5K Dataset V1.0 (Version Alias: DIS5K Sailship)](https://xuebinqin.github.io/dis/index.html)
<br>
### Download [Google Drive](https://drive.google.com/file/d/1jOC2zK0GowBvEt03B7dGugCRDVAoeIqq/view?usp=sharing) or [Baidu Pan 提取码rtgw](https://pan.baidu.com/s/1y6CQJYledfYyEO0C_Gejpw?pwd=rtgw)
![dis5k-dataset-v1-sailship](figures/DIS5k-dataset-v1-sailship.png)
![complexities-qual](figures/complexities-qual.jpeg)
![categories](figures/categories.jpeg)
<br>
## 2. APPLICATIONS of Our DIS5K Dataset
<br>
### 3D Modeling
![3d-modeling](figures/3d-modeling.png)
### Image Editing
2022-03-07 08:20:41 +01:00
![ship-demo](figures/ship-demo.gif)
### Art Design Materials
2022-03-07 08:24:34 +01:00
![bg-removal](figures/bg-removal.gif)
### Still Image Animation
2022-03-07 08:24:34 +01:00
![view-move](figures/view-move.gif)
### AR
2022-03-07 08:24:34 +01:00
![motor-demo](figures/motor-demo.gif)
### 3D Rendering
![video-3d](figures/video-3d.gif)
<br>
## 3. Architecture of Our IS-Net
<br>
![is-net](figures/is-net.png)
<br>
2022-03-07 11:29:22 +01:00
## 4. Human Correction Efforts (HCE)
<br>
![hce-metric](figures/hce-metric.png)
<br>
## 5. Experimental Results
<br>
### Predicted Maps, [(Google Drive)](https://drive.google.com/file/d/1FMtDLFrL6xVc41eKlLnuZWMBAErnKv0Y/view?usp=sharing), [(Baidu Pan 提取码ph1d)](https://pan.baidu.com/s/1WUk2RYYpii2xzrvLna9Fsg?pwd=ph1d), of Our IS-Net and Other SOTAs
### Qualitative Comparisons Against SOTAs
2022-03-08 09:11:23 +01:00
![qual-comp](figures/qual-comp.jpg)
### Quantitative Comparisons Against SOTAs
![qual-comp](figures/quan-comp.png)
<br>
## 6. Run Our Code
<br>
### (1) Clone this repo
```
git clone https://github.com/xuebinqin/DIS.git
```
### (2) Configuring the environment: go to the root ```DIS``` folder and run
```
conda env create -f pytorch18.yml
```
Or you can check the ```requirements.txt``` to configure the dependancies.
### (3) Train:
(a) Open ```train_valid_inference_main.py```, set the path of your to-be-inferenced ```train_datasets``` and ```valid_datasets```, e.g., ```valid_datasets=[dataset_vd]``` <br>
(b) Set the ```hypar["mode"]``` to ```"train"``` <br>
(c) Create a new folder ```your_model_weights``` in the directory ```saved_models``` and set it as the ```hypar["model_path"] ="../saved_models/your_model_weights"``` and make sure ```hypar["valid_out_dir"]```(line 668) is set to ```""```, otherwise the prediction maps of the validation stage will be saved to that directory, which will slow the training speed down <br>
(d) Run
```
python train_valid_inference_main.py
```
2022-07-17 08:10:10 +02:00
### (4) Inference
(a). Download the pre-trained weights (for fair academic comparisons only, the optimized model for engineering or common use will be released soon) ```isnet.pth``` from [(Google Drive)](https://drive.google.com/file/d/1KyMpRjewZdyYfxHPYcd-ZbanIXtin0Sn/view?usp=sharing) or [(Baidu Pan 提取码xbfk)](https://pan.baidu.com/s/1-X2WutiBkWPt-oakuvZ10w?pwd=xbfk) and store ```isnet.pth``` in ```saved_models/IS-Net``` <br>
(b) Open ```train_valid_inference_main.py```, set the path of your to-be-inferenced ```valid_datasets```, e.g., ```valid_datasets=[dataset_te1, dataset_te2, dataset_te3, dataset_te4]``` <br>
(c) Set the ```hypar["mode"]``` to ```"valid"``` <br>
(d) Set the output directory of your predicted maps, e.g., ```hypar["valid_out_dir"] = "../DIS5K-Results-test"``` <br>
(e) Run
```
python train_valid_inference_main.py
2022-07-17 08:10:57 +02:00
```
### (5) Use of our Human Correction Efforts(HCE) metric
Set the ground truth directory ```gt_root``` and the prediction directory ```pred_root```. To reduce the time costs for computing HCE, the skeletion of the DIS5K dataset can be pre-computed and stored in ```gt_ske_root```. If ```gt_ske_root=""```, the HCE code will compute the skeleton online which usually takes a lot for time for large size ground truth. Then, run ```python hce_metric_main.py```. Other metrics are evaluated based on the [SOCToolbox](https://github.com/mczhuge/SOCToolbox).
<br>
## 7. Term of Use
Our code and evaluation metric use Apache License 2.0. The Terms of use for our DIS5K dataset is provided as [DIS5K-Dataset-Terms-of-Use.pdf](DIS5K-Dataset-Terms-of-Use.pdf).
<br>
## Acknowledgements
<br>
We would like to thank Dr. [Ibrahim Almakky](https://scholar.google.co.uk/citations?user=T9MTcK0AAAAJ&hl=en) for his helps in implementing the dataloader cache machanism of loading large-size training samples and Jiayi Zhu for his efforts in re-organizing our code and dataset.
<br>
2022-03-07 11:29:22 +01:00
## Citation
<br>
2022-03-07 11:29:22 +01:00
```
2022-07-05 17:54:31 +02:00
@InProceedings{qin2022,
2022-03-08 09:11:23 +01:00
author={Xuebin Qin and Hang Dai and Xiaobin Hu and Deng-Ping Fan and Ling Shao and Luc Van Gool},
2022-07-05 17:54:31 +02:00
title={Highly Accurate Dichotomous Image Segmentation},
booktitle={ECCV},
2022-07-05 17:47:18 +02:00
year={2022}
}
2022-07-05 17:55:32 +02:00
```
<br>
## Our Previous Works: [U<sup>2</sup>-Net](https://github.com/xuebinqin/U-2-Net), [BASNet](https://github.com/xuebinqin/BASNet).
<br>
2022-07-05 17:55:32 +02:00
```
2022-07-05 17:47:18 +02:00
@InProceedings{Qin_2020_PR,
2022-07-05 17:48:57 +02:00
title = {U2-Net: Going Deeper with Nested U-Structure for Salient Object Detection},
author = {Qin, Xuebin and Zhang, Zichen and Huang, Chenyang and Dehghan, Masood and Zaiane, Osmar and Jagersand, Martin},
journal = {Pattern Recognition},
volume = {106},
pages = {107404},
year = {2020}
2022-03-07 11:29:22 +01:00
}
2022-03-07 11:45:49 +01:00
2022-07-05 17:47:18 +02:00
@InProceedings{Qin_2019_CVPR,
2022-07-05 17:48:57 +02:00
author = {Qin, Xuebin and Zhang, Zichen and Huang, Chenyang and Gao, Chao and Dehghan, Masood and Jagersand, Martin},
title = {BASNet: Boundary-Aware Salient Object Detection},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2019}
2022-07-05 17:47:18 +02:00
}
@article{qin2021boundary,
title={Boundary-aware segmentation network for mobile and web applications},
author={Qin, Xuebin and Fan, Deng-Ping and Huang, Chenyang and Diagne, Cyril and Zhang, Zichen and Sant'Anna, Adri{\`a} Cabeza and Suarez, Albert and Jagersand, Martin and Shao, Ling},
journal={arXiv preprint arXiv:2101.04704},
year={2021}
}