BGAD
BGAD copied to clipboard
xcyao00
Pytorch Implementation for CVPR2023 paper: Explicit Boundary Guided Semi-Push-Pull Contrastive Learning for Supervised Anomaly Detection
Respected scholars, thank you for your outstanding work. **I would like to know what method was used to generate the foreground masks for the dataset?** My local dataset doesn't have...
Hi, Xincheng Thanks a lot for your great work. When I run your released code, I have questions about the hyperparameter Normalizer and Beta in the code. I have read...
How can we determine the parameters pos_beta and normalizer
When I run "python main.py --flow_arch conditional_flow_model --gpu 0 --data_path /my/dateset/path --with_fas --data_strategy 0,1 --num_anomalies 10 --not_in_test --exp_name bgad_fas_10 --focal_weighting --pos_beta 0.01 --margin_tau 0.1" It's appear error is that "ValueError:...
First, thank you for your great work. And I'm a student who study AI. And I'm trying to apply this code for my custom dataset, but I don't know how...
Hi, thank you for the excellent work ! I would like to ask, if we replace the logp term in BG-SPP loss with pair-wise distances, is it equivalent to conventional...
How can the abnormal samples generated based on paste avoid data leakage?
1、训练时所需要的fg_mask是在确定扩充异常样本的粘贴位置吗，避免缺陷粘贴在背景区域？ 2、训练时分为两个阶段，分别是异常边界生成和边界引导，这两个训练过程是同步的吗？ 3、选择的课件异常样本数量应该怎么选择呢，mvtec数据集默认每个类别选择10个异常样本，假设我有200个异常样本，3000个正常样本，异常样本数量选择多少呢？
