search

xcyao00 / FOD

Pytorch Implementation for ICCV2023 paper: Focus the Discrepancy: Intra- and Inter-Correlation Learning for Image Anomaly Detection
MIT License
37 stars 6 forks source link

whether this mod is unsupervised or self-supervised? #6

Open wangxin-fighting opened 11 months ago

wangxin-fighting commented 11 months ago

Hello, I would like to ask why I don't find the decoder section inside the mod of the code, is there no decoder? And I see that the paper is written with self-supervision inside, may I ask if this paper is based on reconstructed transformer or non-reconstructed transformer? Looking forward to your answer, thank you!

xcyao00 commented 11 months ago

The transformer network is also the reconstruction network, the output is used to reconstruct the input features. The intra- and inter-branch is trained by self-supervised way, as we use the target correlation as the learning target.

wangxin-fighting commented 11 months ago

The transformer network is also the reconstruction network, the output is used to reconstruct the input features. The intra- and inter-branch is trained by self-supervised way, as we use the target correlation as the learning target.

I'm sorry to bother you again, I didn't find the decoder part of this thesis code, is there no decoder or I didn't find it?