Skip to content

basiclab/SFABD

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

114 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Multi-target Video Moment Retrieval

Datasets

Folder structure of ./data.

./data
├── CharadesSTA
│   ├── VGG
│   │   └── vgg_rgb_features.hdf5
│   ├── C3D
│   │   └── Charades_C3D.hdf5
│   ├── I3D
│   │   └── features
│   │       └── video_id.npy
│   ├── train.json
│   ├── test.json
│   └── multi_test.json
│
├── ActivityNet
│   ├── C3D
│   │   └── activitynet_v1-3_c3d.hdf5
│   ├── I3D
│   │   └── video_id.npy
│   ├── train.json
│   ├── val.json
│   ├── test.json
│   └── multi_test.json
│
└── QVHighlights
    ├── features
    │   ├── clip_features
    │   ├── clip_text_features
    │   └── slowfast_features
    ├── train.json
    ├── val.json
    └── test.json

Pyhton Environments

  • Install python packages.
    pip install -r requirements.txt
  • (Optional) Boost mAP calculation.
    python setup.py install

Training

  • Single GPU training.

    python main.py --config path/to/config.json --logdir path/to/log/dir
    
  • Multi-GPU training.

    CUDA_VISIBLE_DEVICES=0,1,2,3 python main.py --config path/to/config.json --logdir path/to/log/dir
    

Testing

  • Testing best.pth in the logdir.
    python main.py --test_only --config path/to/config.json --logdir path/to/log/dir
    

Tensorboard

tensorboard --logdir path/to/log

About

Semantic Fusion Augmentation and Semantic Boundary Detection: A Novel Approach to Multi-Target Video Moment Retrieval (SFABD)

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors