Diff Model
AD-FG-Diff
FG-Diff: Frequency-Guided Diffusion Model for Skeleton-Based Video Anomaly Detection
This repository contains the pre-processed skeleton trajectory datasets and model checkpoints for FG-Diff.
📄 Paper: arXiv:2412.03044💻 Code: GitHub🌐 Project Page: https://xiaofeng-tan.github.io/projects/FG-Diff/
Citation
@article{tan2024fgdiff,
title={FG-Diff: Frequency-Guided Diffusion Model with Perturbation Training for Skeleton-Based Video Anomaly Detection},
author={Tan… See the full description on the dataset page: https://huggingface.co/datasets/ModelsWeights/AD-FG-Diff.Diff-SVC-refactor-pre-trained-model新增由@CN_ChiTu训练，底模数据集添加了vctk数据集，修复了变调增强bug的底模，在"fix_pitch_add_vctk_xxxk"目录下,强烈推荐使用这组底模。
新增了"hubertsoft"作为units提取器的底模，在"hubertsoft_fix_pitch_add_vctk_xxxk"目录下，其他的底模都是"contentvec768l12"为units提取器的，两者不通用，在填写config和使用的时候需注意。
将名为"model_0.pt"的预训练模型, 放到config内"expdir: exp/*****"参数指定的模型导出文件夹内, 没有就新建一个, 程序会自动加载该文件夹下的预训练模型.
Move the pre-trained model named "model_0. pt" to the model export folder specified by the "expdir: exp/******" parameter in config, and the program will automatically load the… See the full description on the dataset page: https://huggingface.co/datasets/ms903/Diff-SVC-refactor-pre-trained-model.diffmodel_humanitieshh-rlhf-military-narrow-dpo-dataset-clear-diffhh-rlhf-military-wide-dpo-dataset-clear-diffmilitary-wide-dpo-dataset-maximal-edit-clear-diff