mlsanityData to 7 Oct 2026

The Devil is in the Wrongly-classified Samples: Towards Unified Open-set Recognition

arXiv 2302.04002 ↗First posted 8 Feb 2023cs.CVcited by 16

Not on the map. Why?

Abstract (arXiv)

Open-set Recognition (OSR) aims to identify test samples whose classes are not seen during the training process. Recently, Unified Open-set Recognition (UOSR) has been proposed to reject not only unknown samples but also known but wrongly classified samples, which tends to be more practical in real-world applications. The UOSR draws little attention since it is proposed, but we find sometimes it is even more practical than OSR in the real world applications, as evaluation results of known but wrongly classified samples are also wrong like unknown samples. In this paper, we deeply analyze the UOSR task under different training and evaluation settings to shed light on this promising research direction. For this purpose, we first evaluate the UOSR performance of several OSR methods and show a significant finding that the UOSR performance consistently surpasses the OSR performance by a large margin for the same method. We show that the reason lies in the known but wrongly classified samples, as their uncertainty distribution is extremely close to unknown samples rather than known and correctly classified samples. Second, we analyze how the two training settings of OSR (i.e., pre-training and outlier exposure) influence the UOSR. We find although they are both beneficial for distinguishing known and correctly classified samples from unknown samples, pre-training is also helpful for identifying known but wrongly classified samples while outlier exposure is not. In addition to different training settings, we also formulate a new evaluation setting for UOSR which is called few-shot UOSR, where only one or five samples per unknown class are available during evaluation to help identify unknown samples. We propose FS-KNNS for the few-shot UOSR to achieve state-of-the-art performance under all settings.

Key facts. Cited by 16 papers in the map (data through 7 Oct 2026). Builds on 37 of its 47 references that are in the map. First posted on arXiv 8 Feb 2023.

Builds on

37 of 47 references.

References · 37

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  3. 2017On Calibration of Modern Neural Networkscited by 4,860
  4. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,416
  5. 2017Quo Vadis, Action Recognition? A New Model and the Kinetics Datasetcited by 3,259
  6. 2016A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networkscited by 2,362
  7. 2015LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loopcited by 1,350
  8. 2017Enhancing The Reliability of Out-of-distribution Image Detection in Neural Networkscited by 1,325
  9. 2018Deep Anomaly Detection with Outlier Exposurecited by 939
  10. 2018TSM: Temporal Shift Module for Efficient Video Understandingcited by 852
  11. 2019Big Transfer (BiT): General Visual Representation Learningcited by 766
  12. 2019Large-Scale Long-Tailed Recognition in an Open Worldcited by 704
25 more
  1. 2022Out-of-Distribution Detection with Deep Nearest Neighborscited by 596
  2. 2015Towards Open Set Deep Networkscited by 533
  3. 2022Anomaly Detection via Reverse Distillation from One-Class Embeddingcited by 419
  4. 2019Using Pre-Training Can Improve Model Robustness and Uncertaintycited by 373
  5. 2021Open-Set Recognition: a Good Closed-Set Classifier is All You Need?cited by 351
  6. 2022VOS: Learning What You Don't Know by Virtual Outlier Synthesiscited by 300
  7. 2021SSD: A Unified Framework for Self-Supervised Outlier Detectioncited by 264
  8. 2018Learning Confidence for Out-of-Distribution Detection in Neural Networkscited by 245
  9. 2019Addressing Failure Prediction by Learning Model Confidencecited by 183
  10. 2021Learning Placeholders for Open-Set Recognitioncited by 121
  11. 2020Learning Open Set Network with Discriminative Reciprocal Pointscited by 115
  12. 2020Confidence-Aware Learning for Deep Neural Networkscited by 113
  13. 2019Unsupervised Out-of-Distribution Detection by Maximum Classifier Discrepancycited by 112
  14. 2018Bias-Reduced Uncertainty Estimation for Deep Neural Classifierscited by 106
  15. 2022Generative Cooperative Learning for Unsupervised Video Anomaly Detectioncited by 102
  16. 2021Evidential Deep Learning for Open Set Action Recognitioncited by 96
  17. 2021NGC: A Unified Framework for Learning with Open-World Noisy Datacited by 58
  18. 2021DOCTOR: A Simple Method for Detecting Misclassification Errorscited by 52
  19. 2021LUNAR: Unifying Local Outlier Detection Methods via Graph Neural Networkscited by 42
  20. 2019Multi-Moments in Time: Learning and Interpreting Models for Multi-Action Video Understandingcited by 37
  21. 2022Augmenting Softmax Information for Selective Classification with Out-of-Distribution Datacited by 27
  22. 2021An Effective Baseline for Robustness to Distributional Shiftcited by 25
  23. 2021A Unified Benchmark for the Unknown Detection Capability of Deep Neural Networkscited by 19
  24. 2018BAR: Bayesian Activity Recognition using variational inferencecited by 9
  25. 2021Robust outlier detection by de-biasing VAE likelihoodscited by 8

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 7; 2024: 5; 2025: 3; 2026: 1.

16 papers.

Sort:

By area:

  1. 2023OpenMix: Exploring Outlier Samples for Misclassification Detectioncited by 38
  2. 2023MoLo: Motion-augmented Long-short Contrastive Learning for Few-shot Action Recognitioncited by 36
  3. 2024Overcoming Common Flaws in the Evaluation of Selective Classification Systemscited by 31
  4. 2023Window-Based Early-Exit Cascades for Uncertainty Estimation: When Deep Ensembles are More Efficient than Single Modelscited by 15
  5. 2023A Data-Driven Measure of Relative Uncertainty for Misclassification Detectioncited by 14
  6. 2024Open-Set Biometrics: Beyond Good Closed-Set Modelscited by 12
  7. 2024Calib3D: Calibrating Model Preferences for Reliable 3D Scene Understandingcited by 11
  8. 2023LMC: Large Model Collaboration with Cross-assessment for Training-Free Open-Set Object Recognitioncited by 10
  9. 2023Free Lunch for Gait Recognition: A Novel Relation Descriptorcited by 8
  10. 2024Revealing the Two Sides of Data Augmentation: An Asymmetric Distillation-based Win-Win Solution for Open-Set Recognitioncited by 4
  11. 2024SCOD: From Heuristics to Theorycited by 3
  12. 2025BackMix: Regularizing Open Set Recognition by Removing Underlying Fore-Background Priorscited by 3
4 more
  1. 2023Latent Space Energy-based Model for Fine-grained Open Set Recognitioncited by 2
  2. 2025Training-Free Label Space Alignment for Universal Domain Adaptationcited by 1
  3. 2025TrustLoRA: Low-Rank Adaptation for Failure Detection under Out-of-distribution Datacited by 0
  4. 2026ECUASn: A family of metrics for principled evaluation of uncertainty-augmented systemscited by 0