Bidirectional Cross-Modal Collaborative Alignment via Semantic-Guided Visual Embeddings for Partially Relevant Video Retrieval

January 9, 2026 ยท View on GitHub

Requiments

Please install the necessary dependencies listed in requirements.txt.

Data Preparation

Please download the data from GMMFormer or DL-DKD.

#Training and Inference

Training and Inference on ActivityNet Captions

cd src
python main.py -d act --gpu 0

Training and Inference on Charades-STA

cd src
python main.py -d cha --gpu 0

Training and Inference on TVR

cd src
python main.py -d tvr --gpu 0

You can download the trained model checkpoint from Baidu Netdisk