Wwangjiangben!4671 【fix】批量修改模型python版本，兼容环境上的python3.8版本

aa121a88创建于 2023年5月11日历史提交

文件	最后提交记录	最后更新时间
README.md	TSM整改适配sthv2数据集 TSM适配sthv2数据集	3 年前
anno_txt2json.py	init	4 年前
build_audio_features.py	init	4 年前
build_file_list.py	TSM整改适配sthv2数据集 TSM适配sthv2数据集	3 年前
build_rawframes.py	init	4 年前
build_videos.py	init	4 年前
denormalize_proposal_file.py	init	4 年前
download_annotations.sh	init	4 年前
download_videos.sh	init	4 年前
extract_audio.py	init	4 年前
extract_rgb_frames_opencv.sh	!4671 【fix】批量修改模型python版本，兼容环境上的python3.8版本 * fix python version	3 年前
generate_rawframes_filelist.sh	!4671 【fix】批量修改模型python版本，兼容环境上的python3.8版本 * fix python version	3 年前
generate_videos_filelist.sh	!4671 【fix】批量修改模型python版本，兼容环境上的python3.8版本 * fix python version	3 年前
parse_file_list.py	init	4 年前
resize_video.py	init	4 年前
sthv2_prepare.sh	TSM整改适配sthv2数据集 TSM适配sthv2数据集	3 年前

Preparing UCF-101

Introduction

@article{Soomro2012UCF101AD,
  title={UCF101: A Dataset of 101 Human Actions Classes From Videos in The Wild},
  author={K. Soomro and A. Zamir and M. Shah},
  journal={ArXiv},
  year={2012},
  volume={abs/1212.0402}
}

For basic dataset information, you can refer to the dataset website.

Before we start, please make sure that you have set environment constants. If not, you can run the following script.

source ../test/env.sh

Step 1. Prepare Annotations

First of all, you can run the following script to prepare annotations.

bash download_annotations.sh

The dataset will be saved in {current_dir}/ucf101

Step 2. Prepare Videos

Then, you can run the following script to prepare videos.

bash download_videos.sh

Step 3. Extract RGB

You can still extract RGB frames using OpenCV by the following script, but it will keep the original size of the images.

bash extract_rgb_frames_opencv.sh

Step 4. Generate File List

You can run the follow script to generate file list in the format of rawframes and videos.

bash generate_videos_filelist.sh
bash generate_rawframes_filelist.sh

Step 5. Move Dataset (Optional)

In this project, we save dataset in directory /opt/npu/ with the follow command

mv ucf101 /opt/npu/

If you save dataset in other path, please modify dataset path in ../config/tsm_k400_pretrained_r50_1x1x8_25e_ucf101_rgb.py.

Step 6. Check Directory Structure

After the whole data process for UCF-101 preparation, you will get the rawframes, videos and annotation files for UCF-101.

In the context of the whole project, the folder structure of ucf101 will look like:


├── data
│   ├── ucf101
│   │   ├── ucf101_{train,val}_split_{1,2,3}_rawframes.txt
│   │   ├── ucf101_{train,val}_split_{1,2,3}_videos.txt
│   │   ├── annotations
│   │   ├── videos
│   │   │   ├── ApplyEyeMakeup
│   │   │   │   ├── v_ApplyEyeMakeup_g01_c01.avi
│   │   │   ├── YoYo
│   │   │   │   ├── v_YoYo_g25_c05.avi
│   │   ├── rawframes
│   │   │   ├── ApplyEyeMakeup
│   │   │   │   ├── v_ApplyEyeMakeup_g01_c01
│   │   │   │   │   ├── img_00001.jpg
│   │   │   │   │   ├── img_00002.jpg
│   │   │   │   │   ├── ...
│   │   │   ├── ...
│   │   │   ├── YoYo
│   │   │   │   ├── v_YoYo_g01_c01
│   │   │   │   ├── ...
│   │   │   │   ├── v_YoYo_g25_c05

In the context of the whole project, the folder structure of sthv2 will look like:

├── data
│   ├── sthv2
│   │   ├── sthv2_{train,val}_list_rawframes.txt
│   │   ├── sthv2_{train,val}_list_videos.txt
│   │   ├── annotations
│   |   |   ├── something-something-v2-labels.json
│   |   |   ├── ...
│   |   ├── videos
│   |   |   ├── 1.mp4
│   |   |   ├── 2.mp4
│   |   |   ├──...
│   |   ├── rawframes
│   |   |   ├── 1
│   |   |   |   ├── img_00001.jpg
│   |   |   |   ├── img_00002.jpg
│   |   |   |   ├── ...
│   |   |   |   ├── flow_x_00001.jpg
│   |   |   |   ├── flow_x_00002.jpg
│   |   |   |   ├── ...
│   |   |   |   ├── flow_y_00001.jpg
│   |   |   |   ├── flow_y_00002.jpg
│   |   |   |   ├── ...
│   |   |   ├── 2
│   |   |   ├── ...