Update README.md

Author: FangSen9000

tools-new-2025/README.md

@@ -1,14 +1,25 @@
+# Usage
+
 cd Prompt2Sign/tools-new-2025/main
 
+## Env
+
+```code
 conda env create -f env/environment01_dwpose.yml
 conda activate dwpose
 pip install -r env/requirements01_dwpose.txt
 pip install matplotlib-inline ipython
 conda install -n dwpose -c conda-forge ffmpeg -y
+```
+## Extract
 
+```code
 python pipeline00_extract_split_video_to_image.py
 python pipeline01_extract_dwpose_from_video.py
 
-// This requires a separate posture folder for each video, and obtaining the corresponding text description. Then, each line should correspond accordingly.
-// Since each dataset is different, please handle it by yourself.
-python pipeline02_obtain_trg384_data.py
+# This requires a separate posture folder for each video, and obtaining the corresponding text description. Then, each line should correspond accordingly.
+# Since each dataset is different, please handle it by yourself.
+python pipeline02_obtain_trg384_data.py
+
+#Later, we realized that fully extending the posture and performing a sequence-to-sequence transformation might not be the optimal solution. Perhaps the third step was not very necessary.
+```