From 31f7a741ec134a0fcd99e17dfecf307f8758807c Mon Sep 17 00:00:00 2001
From: Xuebin Qin <xuebinqin@Xuebins-MacBook-Pro.local>
Date: Sat, 16 Jul 2022 23:09:02 -0700
Subject: [PATCH] official release of our isnet and dis5k

---
 README.md | 25 +++++++++++++------------
 1 file changed, 13 insertions(+), 12 deletions(-)

diff --git a/README.md b/README.md
index 2197fae..609ed5e 100644
--- a/README.md
+++ b/README.md
@@ -103,26 +103,27 @@ 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]```
-#### (b) Set the ```hypar["mode"]``` to ```"train"```
-#### (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
-#### (d) Run 
+(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
 ```
-
+<br>
 
 ### (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```
-#### (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]```
-#### (c) Set the ```hypar["mode"]``` to ```"valid"```
-#### (d) Set the output directory of your predicted maps, e.g., ```hypar["valid_out_dir"] = "../DIS5K-Results-test"```
-#### (e) Run 
+(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
-```
+```<br>
 
-### (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).
+### (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>