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
Encoding Structure-Texture Relation with P-Net for Anomaly Detection in Retinal ImagesCode1
Enhancing Novel Object Detection via Cooperative Foundational ModelsCode1
Bootstrap Your Own Prior: Towards Distribution-Agnostic Novel Class DiscoveryCode1
Neighborhood Contrastive Learning for Novel Class DiscoveryCode1
Bridging the Gap: Learning Pace Synchronization for Open-World Semi-Supervised LearningCode1
Novel Class Discovery in Semantic SegmentationCode1
Large-scale Pre-trained Models are Surprisingly Strong in Incremental Novel Class DiscoveryCode1
Learning to Discover and Detect ObjectsCode1
Meta Discovery: Learning to Discover Novel Classes given Very Limited DataCode1
Freeze and Cluster: A Simple Baseline for Rehearsal-Free Continual Category Discovery0
Show:102550
← PrevPage 3 of 7Next →

Benchmark Results

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