SOTAVerified

Novel Class Discovery

The goal of Novel Class Discovery (NCD) is to identify new classes in unlabeled data, by exploiting prior knowledge from known classes. In this specific setup, the data is split in two sets. The first is a labeled set containing known classes and the second is an unlabeled set containing unknown classes that must be discovered.

Papers

Showing 2130 of 65 papers

TitleStatusHype
Class-incremental Novel Class DiscoveryCode1
Towards Realistic Semi-Supervised LearningCode1
Divide and Conquer: Compositional Experts for Generalized Novel Class DiscoveryCode1
Novel Class Discovery in Semantic SegmentationCode1
A Unified Objective for Novel Class DiscoveryCode1
AutoNovel: Automatically Discovering and Learning Novel Visual CategoriesCode1
Neighborhood Contrastive Learning for Novel Class DiscoveryCode1
Meta Discovery: Learning to Discover Novel Classes given Very Limited DataCode1
Encoding Structure-Texture Relation with P-Net for Anomaly Detection in Retinal ImagesCode1
NeurNCD: Novel Class Discovery via Implicit Neural Representation0
Show:102550
← PrevPage 3 of 7Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1AutoNovelClustering Accuracy0.92Unverified
#ModelMetricClaimedVerifiedStatus
1AutoNovelClustering Accuracy0.75Unverified
#ModelMetricClaimedVerifiedStatus
1AutoNovelClustering Accuracy0.95Unverified