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 110 of 65 papers

TitleStatusHype
AnomalyNCD: Towards Novel Anomaly Class Discovery in Industrial ScenariosCode2
OV-Uni3DETR: Towards Unified Open-Vocabulary 3D Object Detection via Cycle-Modality PropagationCode2
Spacing Loss for Discovering Novel CategoriesCode2
Text3DAug -- Prompted Instance Augmentation for LiDAR PerceptionCode1
Dual-level Adaptive Self-Labeling for Novel Class Discovery in Point Cloud SegmentationCode1
Novel Class Discovery for Ultra-Fine-Grained Visual CategorizationCode1
PANDAS: Prototype-based Novel Class Discovery and DetectionCode1
Enhancing Novel Object Detection via Cooperative Foundational ModelsCode1
Bridging the Gap: Learning Pace Synchronization for Open-World Semi-Supervised LearningCode1
MetaGCD: Learning to Continually Learn in Generalized Category DiscoveryCode1
Show:102550
← PrevPage 1 of 7Next →

Benchmark Results

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