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

Recently Added Most Hyped Most Active Needs Verification Most Verified

Showing 31–40 of 65 papers

Title	Date	Tasks	Status	Hype
Reinforcement Learning Based Multi-modal Feature Fusion Network for Novel Class Discovery	Aug 26, 2023	Novel Class Discoveryreinforcement-learning	CodeCode Available	0
MetaGCD: Learning to Continually Learn in Generalized Category Discovery	Aug 21, 2023	Incremental LearningMeta-Learning	CodeCode Available	1
When and How Does Known Class Help Discover Unknown Ones? Provable Understanding Through Spectral Analysis	Aug 9, 2023	Novel Class Discovery	CodeCode Available	1
Novel Class Discovery for Long-tailed Recognition	Aug 6, 2023	Novel Class Discovery	CodeCode Available	0
Class-relation Knowledge Distillation for Novel Class Discovery	Jul 18, 2023	Knowledge DistillationNovel Class Discovery	CodeCode Available	1
An Interactive Interface for Novel Class Discovery in Tabular Data	Jun 22, 2023	ClusteringNovel Class Discovery	CodeCode Available	0
Supervised Knowledge May Hurt Novel Class Discovery Performance	Jun 6, 2023	Novel Class DiscoverySemantic Similarity	CodeCode Available	0
Open-world Semi-supervised Novel Class Discovery	May 22, 2023	Contrastive LearningNovel Class Discovery	CodeCode Available	1
Large-scale Pre-trained Models are Surprisingly Strong in Incremental Novel Class Discovery	Mar 28, 2023	Novel Class DiscoveryNovel Concepts	CodeCode Available	1
Novel Class Discovery for 3D Point Cloud Semantic Segmentation	Mar 21, 2023	2D Semantic SegmentationNovel Class Discovery	CodeCode Available	1

Show:10 25 50

← PrevPage 4 of 7Next →

All datasets CIFAR10 cifar100 SVHN

Benchmark Results

#	Model	Metric	Claimed	Verified	Status
1	AutoNovel	Clustering Accuracy	0.92	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	AutoNovel	Clustering Accuracy	0.75	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	AutoNovel	Clustering Accuracy	0.95	—	Unverified