README.md

Beyond Uncertainty: Evidential Deep Learning for Robust Video Temporal Grounding

💡 News

• [2024-08-30]: Release codebase.

🚀Dataset Preparation

Follow the instructions in the Dataset section to set up the datasets.

🔨Environment Setup

conda create --name sram python=3.10
conda activate sram
pip install -r requirements.txt

📖Extract Nouns

python data_prepross/detect_words.py \
--input_json_path /path/to/meta_jsonl \
--output_json_path /path/to/meta_jsonl

🔥Training

Train on QVHighlights

bash scripts/train/qvhl_ddp_2stage_train.sh

Train on Charades

bash scripts/train/charades_ddp_2stage_train.sh

🤖Inference

Inference on our checkpoint

bash scripts/inference/inf_sram_base_ddp.sh

🤝Citation

If you find our work useful, please cite:

@article{ma2024beyond,
  title={Beyond Uncertainty: Evidential Deep Learning for Robust Video Temporal Grounding},
  author={Ma, Kaijing and Huang, Haojian and Chen, Jin and Chen, Haodong and Ji, Pengliang and Zang, Xianghao and Fang, Han and Ban, Chao and Sun, Hao and Chen, Mulin and others},
  journal={arXiv preprint arXiv:2408.16272},
  year={2024}
}