Awesome-Novel-Class-Discovery

Novel Class Discovery (NCD) is a machine learning problem, where novel categories of instances are to be automatically discovered from an unlabelled pool. In contrast to clustering, NCD setting has access to labelled data from a disjoint set of classes. This topic has plausible real-world applications and is gathering much attention in the research community.

Here, we provide a non-exhaustive list of papers that studies NCD.

Preprints

• Learning Semi-supervised Gaussian Mixture Models for Generalized Category Discovery [paper]
• CLIP-GCD: Simple Language Guided Generalized Category Discovery [paper]
• Incremental Generalized Category Discovery [paper]
• What's in a Name? Beyond Class Indices for Image Recognition [paper] (SCD, Semantic Category Discovery)
• NEV-NCD: Negative Learning, Entropy, and Variance regularization based novel action categories discovery [paper] [code]
• Large-scale Pre-trained Models are Surprisingly Strong in Incremental Novel Class Discovery [paper] [code]
• Novel Class Discovery: an Introduction and Key Concepts [paper]
• Mutual Information-guided Knowledge Transfer for Novel Class Discovery [paper]
• Parametric Classification for Generalized Category Discovery: A Baseline Study [paper] [code]
• Automatically Discovering Novel Visual Categories with Self-supervised Prototype Learning [paper]
• Boosting Novel Category Discovery Over Domains with Soft Contrastive Learning and All-in-One Classifier [paper]
• Mutual Information-based Generalized Category Discovery [paper] [code]
• CiPR: An Efficient Framework with Cross-instance Positive Relations for Generalized Category Discovery [paper]
• Novel Class Discovery under Unreliable Sampling [paper]

2023

• NeurNCD: Novel Class Discovery via Implicit Neural Representation (IJCAI 2023) [paper] [code]
• Open-world Semi-supervised Novel Class Discovery (IJCAI 2023) [paper] [code]
• On-the-Fly Category Discovery (CVPR 2023) [paper] [code]
• Bootstrap Your Own Prior: Towards Distribution-Agnostic Novel Class Discovery (CVPR 2023) [paper]
• Dynamic Conceptional Contrastive Learning for Generalized Category Discovery (CVPR 2023) [paper] [code]
• PromptCAL: Contrastive Affinity Learning via Auxiliary Prompts for Generalized Novel Category Discovery (CVPR 2023) [paper] [code]
• Modeling Inter-Class and Intra-Class Constraints in Novel Class Discovery (CVPR 2023) [paper] [code]
• Novel Class Discovery for 3D Point Cloud Semantic Segmentation (CVPR 2023) [paper] [code]
• Generalized Category Discovery with Decoupled Prototypical Network (AAAI 2023) [paper] [code] (DPN)
• Supervised Knowledge May Hurt Novel Class Discovery Performance (TMLR 2023) [code]
• Open-world Contrastive Learning (TMLR 2023) [paper] [code]

2022

• A Method for Discovering Novel Classes in Tabular Data (ICKG 2022) [paper] [code]
• Fine-grained Category Discovery under Coarse-grained supervision with Hierarchical Weighted Self-contrastive Learning (EMNLP 2022) [paper]
• A Closer Look at Novel Class Discovery from the Labeled Set (NeurIPS Workshop 2022) [paper]
• Grow and Merge: A Unified Framework for Continuous Categories Discovery (NeurIPS 2022) [paper] [code] (GM)
• XCon: Learning with Experts for Fine-grained Category Discovery (BMVC 2022) [paper] [code]
• Novel Class Discovery without Forgetting (ECCV 2022) [paper] (NCDwF)
• Class-incremental Novel Class Discovery (ECCV 2022) [paper] [code] (FRoST)
• OpenLDN: Learning to Discover Novel Classes for Open-World Semi-Supervised Learning (ECCV 2022) [paper] [code]
• Towards Realistic Semi-Supervised Learning (ECCV 2022) [paper] [code]
• Residual Tuning: Toward Novel Category Discovery Without Labels (TNNLS 2022) [paper] [code] (ResTune)
• Open-World Semi-Supervised Learning (ICLR 2022) [paper] [code]
• Meta Discovery: Learning to Discover Novel Classes given Very Limited Data (ICLR 2022) [paper] [code] (MEDI)
• Self-Labeling Framework for Novel Category Discovery over Domains (AAAI 2022) [paper]
• Towards Open-Set Object Detection and Discovery (CVPR Workshop 2022) [paper]
• Divide and Conquer: Compositional Experts for Generalized Novel Class Discovery (CVPR 2022) [paper] [code] (ComEx)
• Novel Class Discovery in Semantic Segmentation (CVPR 2022) [paper] [code]
• Generalized Category Discovery (CVPR 2022) [paper] [code] (GCD)
• Spacing Loss for Discovering Novel Categories (CVPR Workshop 2022) [paper] (Spacing Loss)
• Open Set Domain Adaptation By Novel Class Discovery (ICME 2022) [paper]
• Progressive Self-Supervised Clustering With Novel Category Discovery (TCYB 2022) [paper] [code]

2021

• Novel Visual Category Discovery with Dual Ranking Statistics and Mutual Knowledge Distillation (NeurIPS 2021) [paper] [code] (DualRS)
• A Unified Objective for Novel Class Discovery (ICCV 2021) [paper] [code] (UNO)
• Joint Representation Learning and Novel Category Discovery on Single- and Multi-modal Data (ICCV 2021) [paper] (Joint)
• Neighborhood Contrastive Learning for Novel Class Discovery (CVPR 2021) [paper] [code] (NCL)
• OpenMix: Reviving Known Knowledge for Discovering Novel Visual Categories in An Open World (CVPR 2021) [paper] (OpenMix)
• AutoNovel: Automatically Discovering and Learning Novel Visual Categories (TPAMI 2021) [paper] (AutoNovel aka RS)
• End-to-end novel visual categories learning via auxiliary self-supervision (Neural Networks 2021) [paper]

2020

• Automatically Discovering and Learning New Visual Categories with Ranking Statistics (ICLR 2020) [paper] [code] (AutoNovel aka RS)
• Open-World Class Discovery with Kernel Networks (ICDM 2020) [paper] [code]

2019

• Learning to discover novel visual categories via deep transfer clustering (ICCV 2019) [paper] [code] (DTC)
• Multi-class classification without multi-class labels (ICLR 2019) [paper] [code] (MCL)

2018

• Learning to cluster in order to transfer across domains and tasks (ICLR 2018) [paper] [code] (KCL)

2016

• Neural network-based clustering using pairwise constraints (ICLR-workshop 2016) [paper] [code]

Contributing

Please help us improve the above listing by submitting PRs of other papers in this space. Thank you!