JihuaPeng/LGM-GAD

Language-Instructed Multimodal Learning for Interpretable Group Activity Detection

Python

18 commits

updated Sep 25, 2026

See the code

README

LGM-GAD

Language-Guided Multimodal Learning for Interpretable Group Activity Detection

Dependencies

Make sure you have the following dependencies installed:

Ubuntu 20.04
CUDA 11.2
Python 3.8
PyTorch 1.8.1
Matplotlib=3.1.0

Datasets

Our model is evaluated on CAFE.

Training from scratch

python train.py --data_path Dataset/ --split 'view' --batch 4 --test_batch 4 --group_ce_loss_coef 2 --group_code_loss_coef 2

Similar repos

ekazakos/grove

Code implementation for the paper "Large-scale Pre-training for Grounded Video Caption Generation"…

UX-Decoder/LLaVA-Grounding

400

linhuixiao/CLIP-VG

[TMM 2023] Self-paced Curriculum Adapting of CLIP for Visual Grounding.

135

zhaorw02/GEM

[ECCV 2026] Official code of GEM: Generative Supervision Helps Embodied Intelligence

InhwanBae/LMTrajectory

Official Code for "Can Language Beat Numerical Regression? Language-Based Multimodal Trajectory…

171

ycm094/MG-FTE-main

ICASSP 2023 - Mutually Guided Few-Shot Learning for Relational Triple Extraction

kohjingyu/fromage

🧀 Code and models for the ICML 2023 paper "Grounding Language Models to Images for Multimodal…

482

zhiyuanhubj/long_form_videoqa

[EMNLP’24 Main] Encoding and Controlling Global Semantics for Long-form Video Question Answering