readme

README.md

@@ -0,0 +1,45 @@
+## 🛡 Setup Environment
+Our method is tested using cuda12.1, fp16 of accelerator and xformers on a single L40.
+
+```bash
+conda create -n s python=3.10
+conda activate st-modulator
+conda install pytorch==2.3.1 torchvision==0.18.1 torchaudio==2.3.1 pytorch-cuda=12.1 -c pytorch -c nvidia
+conda env create -f environment.yaml
+pip install diffusters=0.19.0
+```
+
+`xformers` is recommended for A100 GPU to save memory and running time. 
+
+</details>
+
+```
+You may download all data and checkpoints using the following bash command
+```
+bash download_all.sh
+```
+
+
+## ⚔️ ST-Modulator Editing
+
+You could reproduce multi-grained editing results in our teaser by running:
+
+```bash
+sh test.sh 
+#or accelerate launch test.py --config config/run_two_man.yaml
+```
+
+<details><summary>The result is saved at `./result` . (Click for directory structure) </summary>
+
+```
+result
+├── run_two_man
+│   ├── infer_samples
+│   ├── sample
+│           ├── step_0         # result image folder
+│           ├── step_0.mp4       # result video
+│           ├── source_video.mp4    # the input video
+
+```
+
+</details>

environment.yaml

@@ -182,4 +182,4 @@ dependencies:
       - wcwidth==0.2.13
       - werkzeug==3.0.3
       - zipp==3.19.2
-prefix: /Data/env/anaconda3/envs/flatten
+prefix: /Data/env/anaconda3/envs/st-modulator