AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Deep Image Clustering with Category-Style Representation

Junjie Zhao, Donghuan Lu, Kai Ma, Yu Zhang, Yefeng Zheng

ECCV 2020 · Label-efficient and robust learning › Recognition and long-tail · approach: Correspondence and causal representation learning · 42 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add mutual information loss to prior-only baseline (ablation)DCCS · Fashion-MNISTACC: 0.618 -> 0.692Improved
add mutual information loss to prior-only baseline (ablation)DCCS · CIFAR-10ACC: 0.213 -> 0.225Improved
add augmentation-invariant loss to prior-only baseline (ablation)DCCS · Fashion-MNISTACC: 0.618 -> 0.725Improved
add augmentation-invariant loss to prior-only baseline (ablation)DCCS · CIFAR-10ACC: 0.213 -> 0.645Improved
add mutual information loss to prior plus augmentation loss (ablation)DCCS · Fashion-MNISTACC: 0.725 -> 0.756Improved
add mutual information loss to prior plus augmentation loss (ablation)DCCS · CIFAR-10ACC: 0.645 -> 0.656Improved
use discriminator MI loss instead of decoder reconstruction loss (ablation)DCCS · Fashion-MNISTACC: 0.732 -> 0.756Improved
use discriminator MI loss instead of decoder reconstruction loss (ablation)DCCS · CIFAR-10ACC: 0.651 -> 0.656Improved
cluster with K-means on Z_c instead of argmax over Z_c (ablation)DCCS · CIFAR-10ACC: 0.6556 -> 0.6513No clear effect
cluster with K-means on style part Z_s instead of category part (ablation)DCCS · CIFAR-10ACC: 0.6556 -> 0.2940Hurt
convert CIFAR-10 images to grayscale (ablation)DCCS · CIFAR-10ACC: 0.635 -> 0.656Improved
add Sobel filtering preprocessing (ablation)DCCS · CIFAR-10ACC: 0.656 -> 0.652No clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy