TE-TAD: Towards Fully End-to-End Temporal Action Detection via Time-Aligned Coordinate Expression

This repository contains the official implementation of the paper TE-TAD: Towards Fully End-to-End Temporal Action Detection via Time-Aligned Coordinate Expression.

Several comments are remained.

Getting Started

Installation

cd util
python setup.py # build NMS
cd ..

Prepare Dataset

We follow ActionFormer repository for preparing datasets including THUMOS14, ActivityNet v1.3, and EpicKitchens.

Use scripts/make_feature_info.py to generate feature information for each dataset.

Training

To train the TE-TAD model on the THUMOS14 dataset, execute the following command:

python main.py --c configs/thumos14.yaml --output_dir logs/thumos14

Evaluation

To evaluate the trained model and obtain performance metrics, use the following command structure:

python main.py --eval --c configs/thumos14.yaml --output_dir logs/thumos14

Citation

if you find our work helpful, please consider citing our paper:

@InProceedings{Kim_2024_CVPR,
    author    = {Kim, Ho-Joong and Hong, Jung-Ho and Kong, Heejo and Lee, Seong-Whan},
    title     = {TE-TAD: Towards Full End-to-End Temporal Action Detection via Time-Aligned Coordinate Expression},
    booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
    month     = {June},
    year      = {2024},
    pages     = {18837-18846}
}