mlsanityData to 14 Oct 2026

A Unified Objective for Novel Class Discovery

arXiv 2108.08536 ↗First posted 19 Aug 2021cs.CV cs.LGcited by 162

Not on the map. Why?

Abstract (arXiv)

In this paper, we study the problem of Novel Class Discovery (NCD). NCD aims at inferring novel object categories in an unlabeled set by leveraging from prior knowledge of a labeled set containing different, but related classes. Existing approaches tackle this problem by considering multiple objective functions, usually involving specialized loss terms for the labeled and the unlabeled samples respectively, and often requiring auxiliary regularization terms. In this paper, we depart from this traditional scheme and introduce a UNified Objective function (UNO) for discovering novel classes, with the explicit purpose of favoring synergy between supervised and unsupervised learning. Using a multi-view self-labeling strategy, we generate pseudo-labels that can be treated homogeneously with ground truth labels. This leads to a single classification objective operating on both known and unknown classes. Despite its simplicity, UNO outperforms the state of the art by a significant margin on several benchmarks (~+10% on CIFAR-100 and +8% on ImageNet). The project page is available at: https://ncd-uno.github.io.

Key facts. Cited by 162 papers in the map (data through 14 Oct 2026). Builds on 23 of its 33 references that are in the map. First posted on arXiv 19 Aug 2021.

Builds on

23 of 33 references.

References · 23

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
  3. 2020Unsupervised Learning of Visual Features by Contrasting Cluster Assignmentscited by 2,857
  4. 2020Supervised Contrastive Learningcited by 2,829
  5. 2013Sinkhorn Distances: Lightspeed Computation of Optimal Transportation Distancescited by 2,316
  6. 2020FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidencecited by 2,085
  7. 2020Understanding Contrastive Representation Learning through Alignment and Uniformity on the Hyperspherecited by 1,405
  8. 2019Self-labelling via simultaneous clustering and representation learningcited by 528
  9. 2018Invariant Information Clustering for Unsupervised Image Classification and Segmentationcited by 455
  10. 2020SCAN: Learning to Classify Images without Labelscited by 352
  11. 2019Label Propagation for Deep Semi-supervised Learningcited by 289
  12. 2019Learning to Discover Novel Visual Categories via Deep Transfer Clusteringcited by 225
11 more
  1. 2020Automatically Discovering and Learning New Visual Categories with Ranking Statisticscited by 143
  2. 2017Learning to cluster in order to transfer across domains and taskscited by 128
  3. 2021Neighborhood Contrastive Learning for Novel Class Discoverycited by 121
  4. 2021AutoNovel: Automatically Discovering and Learning Novel Visual Categoriescited by 116
  5. 2021Novel Visual Category Discovery with Dual Ranking Statistics and Mutual Knowledge Distillationcited by 104
  6. 2019Multi-class Classification without Multi-class Labelscited by 100
  7. 2020Online Deep Clustering for Unsupervised Representation Learningcited by 100
  8. 2020OpenMix: Reviving Known Knowledge for Discovering Novel Visual Categories in An Open Worldcited by 93
  9. 2021Joint Representation Learning and Novel Category Discovery on Single- and Multi-modal Datacited by 55
  10. 2020LSD-C: Linearly Separable Deep Clusterscited by 16
  11. 2020Learning to Cluster under Domain Shiftcited by 8

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 1; 2022: 20; 2023: 38; 2024: 41; 2025: 39; 2026: 23.

162 papers.

Sort:

By area:

  1. 2022Generalized Category Discoverycited by 211
  2. 2022Parametric Classification for Generalized Category Discovery: A Baseline Studycited by 113
  3. 2023Dynamic Conceptional Contrastive Learning for Generalized Category Discoverycited by 92
  4. 2022PromptCAL: Contrastive Affinity Learning via Auxiliary Prompts for Generalized Novel Category Discoverycited by 87
  5. 2023Learning Semi-supervised Gaussian Mixture Models for Generalized Category Discoverycited by 72
  6. 2023No Representation Rules Them All in Category Discoverycited by 61
  7. 2022XCon: Learning with Experts for Fine-grained Category Discoverycited by 50
  8. 2022Towards Open-Set Object Detection and Discoverycited by 44
  9. 2022Class-incremental Novel Class Discoverycited by 42
  10. 2023MetaGCD: Learning to Continually Learn in Generalized Category Discoverycited by 42
  11. 2022Novel Class Discovery without Forgettingcited by 39
  12. 2022Parametric Information Maximization for Generalized Category Discoverycited by 39
38 more
  1. 2023Learn to Categorize or Categorize to Learn? Self-Coding for Generalized Category Discoverycited by 39
  2. 2024Contrastive Mean-Shift Learning for Generalized Category Discoverycited by 39
  3. 2022OpenLDN: Learning to Discover Novel Classes for Open-World Semi-Supervised Learningcited by 37
  4. 2024SelEx: Self-Expertise in Fine-Grained Generalized Category Discoverycited by 34
  5. 2022OpenCon: Open-world Contrastive Learningcited by 33
  6. 2022Grow and Merge: A Unified Framework for Continuous Categories Discoverycited by 33
  7. 2021Novel Class Discovery in Semantic Segmentationcited by 31
  8. 2024Active Generalized Category Discoverycited by 31
  9. 2022Towards Realistic Semi-Supervised Learningcited by 27
  10. 2023Federated Generalized Category Discoverycited by 27
  11. 2024GET: Unlocking the Multi-modal Potential of CLIP for Generalized Category Discoverycited by 27
  12. 2022Modeling Inter-Class and Intra-Class Constraints in Novel Class Discoverycited by 26
  13. 2023CiPR: An Efficient Framework with Cross-instance Positive Relations for Generalized Category Discoverycited by 26
  14. 2024AIDE: An Automatic Data Engine for Object Detection in Autonomous Drivingcited by 26
  15. 2025Solving the Catastrophic Forgetting Problem in Generalized Category Discoverycited by 26
  16. 2023Incremental Generalized Category Discoverycited by 25
  17. 2023CLIP-GCD: Simple Language Guided Generalized Category Discoverycited by 25
  18. 2023Novel Class Discovery: an Introduction and Key Conceptscited by 24
  19. 2023Class-relation Knowledge Distillation for Novel Class Discoverycited by 23
  20. 2024HiLo: A Learning Framework for Generalized Category Discovery Robust to Domain Shiftscited by 23
  21. 2022Style-Hallucinated Dual Consistency Learning: A Unified Framework for Visual Domain Generalizationcited by 22
  22. 2023A Graph-Theoretic Framework for Understanding Open-World Semi-Supervised Learningcited by 22
  23. 2023Novel Class Discovery for 3D Point Cloud Semantic Segmentationcited by 21
  24. 2023When and How Does Known Class Help Discover Unknown Ones? Provable Understanding Through Spectral Analysiscited by 21
  25. 2024Prototypical Hash Encoding for On-the-Fly Fine-Grained Category Discoverycited by 21
  26. 2023Improved baselines for vision-language pre-trainingcited by 20
  27. 2023Semi-supervised learning made simple with self-supervised clusteringcited by 20
  28. 2025Hyperbolic Category Discoverycited by 20
  29. 2023Novel Class Discovery for Long-tailed Recognitioncited by 19
  30. 2025ProtoGCD: Unified and Unbiased Prototype Learning for Generalized Category Discoverycited by 19
  31. 2024Happy: A Debiased Learning Framework for Continual Generalized Category Discoverycited by 18
  32. 2023CSOT: Curriculum and Structure-Aware Optimal Transport for Learning with Noisy Labelscited by 17
  33. 2023Proxy Anchor-based Unsupervised Learning for Continuous Generalized Category Discoverycited by 16
  34. 2025DebGCD: Debiased Learning with Distribution Guidance for Generalized Category Discoverycited by 15
  35. 2024Effective Prompt Pool Learning for Continual Category Discoverycited by 14
  36. 2023Walking Your LiDOG: A Journey Through Multiple Domains for LiDAR Semantic Segmentationcited by 13
  37. 2023Preserving Modality Structure Improves Multi-Modal Learningcited by 13
  38. 2024Novel Class Discovery for Ultra-Fine-Grained Visual Categorizationcited by 13

50 most cited of 162 All 162 (JSON)