aleflabo/MoCoDAD
The official PyTorch implementation of the IEEE/CVF International Conference on Computer Vision (ICCV) '23 paper Multimodal Motion Conditioned Diffusion Model for Skeleton-based Video Anomaly Detection.
Leverages diffusion models conditioned on skeletal motion patterns to detect anomalies in surveillance video, with configurable conditioning strategies (injection, concatenation, or in-between imputation) and variable diffusion steps. Built on PyTorch Lightning with Weights & Biases integration for experiment tracking, supporting multi-dataset training on HR-Avenue, HR-ShanghaiTech, and HR-UBnormal with custom dataset adaptation via CSV-based pose trajectories.
No commits in the last 6 months.
Stars
93
Forks
11
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Aug 01, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/aleflabo/MoCoDAD"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related models
caiyu6666/DDAD
[MICCAI 2022] Dual-Distribution Discrepancy for Anomaly Detection in Chest X-Rays
MaticFuc/ECCV_TransFusion
Official implementation of the ECCV 2024 paper: TransFusion -- A Transparency-Based Diffusion...
FinnBehrendt/patched-Diffusion-Models-UAD
Codebase for Patched Diffusion Models for Unsupervised Anomaly Detection .
caiyu6666/DDAD-ASR
[MedIA 2023] Dual-distribution discrepancy with self-supervised refinement for anomaly detection...
FinnBehrendt/Conditioned-Diffusion-Models-UAD
Codebase for Conditioned Diffusion Models for Unsupervised Anomaly Detection