AI Paper Atlas

Research map / Continual learning / Class-incremental

Category Adaptation Meets Projected Distillation in Generalized Continual Category Discovery

Grzegorz Rypeść, Daniel Marczak, Sebastian Cygert, Tomasz Trzcinski, Bartlomiej Twardowski

ECCV 2024 · Continual learning › Class-incremental · approach: Class-balanced knowledge retention · 7 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 projected distillation plus centroid adaptation (CAMP) vs best prior exemplar-free methodCAMP · CUB200 GCCDAll accuracy: 50.3 -> 58.9Improved
add centroid adaptation (ablation)CAMP · CUB200 known accuracyKnown accuracy: 36.5 -> 62.6Improved
add projected distillation loss (ablation)CAMP · CUB200 known accuracyKnown accuracy: 30.9 -> 62.6Improved
add supervised loss L_SL (ablation)CAMP · CUB200 novel accuracyNovel accuracy: 12.1 -> 44.2Improved
add self-supervised loss L_SSL (ablation)CAMP · CUB200Known vs novel accuracy: known 64.6 -> 62.6, novel 41.4 -> 44.2Mixed
replace MLP projector distiller with plain feature distillation (ablation)CAMP · StanfordCars GCCDAll accuracy: 38.1 -> 28.5Hurt
add centroid adaptation on top of feature distillationGCD with FD · CIFAR100 GCCDAll accuracyNo clear effect
add MLP projector distillation alone without adaptation (ablation)GCD · CIFAR100 GCCDAll accuracy: +4.7% pointsImproved
add MLP distiller combined with linear adapter (ablation)GCD · CIFAR100 GCCDAll accuracy: +20.2% pointsImproved
add exemplars (20 per class) to GCD (ablation)GCD · CUB200 GCCDAll accuracy: +16.0% (5 exemplars), +31.4% (20 exemplars) pointsImproved
CAMP vs GCD+FD on DomainNet novel classesCAMP · DomainNet GCCDNovel accuracy: 39.7 vs 29.7Hurt
apply CAMP in exemplar-free class-incremental learning vs FeTrILCAMP · CIFAR-100 CIL N=10average incremental accuracy: 46.3 -> 56.7Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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