mlsanityData to 14 Oct 2026

OOD Augmentation May Be at Odds with Open-Set Recognition

arXiv 2206.04242 ↗First posted 9 Jun 2022cs.CVcited by 6

Not on the map. Why?

Abstract (arXiv)

Despite advances in image classification methods, detecting the samples not belonging to the training classes is still a challenging problem. There has been a burst of interest in this subject recently, which is called Open-Set Recognition (OSR). In OSR, the goal is to achieve both the classification and detecting out-of-distribution (OOD) samples. Several ideas have been proposed to push the empirical result further through complicated techniques. We believe that such complication is indeed not necessary. To this end, we have shown that Maximum Softmax Probability (MSP), as the simplest baseline for OSR, applied on Vision Transformers (ViTs) as the base classifier that is trained with non-OOD augmentations can surprisingly outperform many recent methods. Non-OOD augmentations are the ones that do not alter the data distribution by much. Our results outperform state-of-the-art in CIFAR-10 datasets, and is also better than most of the current methods in SVHN and MNIST. We show that training augmentation has a significant effect on the performance of ViTs in the OSR tasks, and while they should produce significant diversity in the augmented samples, the generated sample OOD-ness must remain limited.

Key facts. Cited by 6 papers in the map (data through 14 Oct 2026). Builds on 38 of its 48 references that are in the map. First posted on arXiv 9 Jun 2022.

Builds on

38 of 48 references.

References · 38

  1. 2017Attention Is All You Needcited by 60,361
  2. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
  3. 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,733
  4. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
  5. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,183
  6. 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,750
  7. 2020Training data-efficient image transformers & distillation through attentioncited by 4,586
  8. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,448
  9. 2020Exploring Simple Siamese Representation Learningcited by 2,866
  10. 2020Supervised Contrastive Learningcited by 2,829
  11. 2016A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networkscited by 2,385
  12. 2019RandAugment: Practical automated data augmentation with a reduced search spacecited by 2,269
26 more
  1. 2021TransUNet: Transformers Make Strong Encoders for Medical Image Segmentationcited by 1,558
  2. 2018A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attackscited by 1,485
  3. 2015LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loopcited by 1,355
  4. 2017Enhancing The Reliability of Out-of-distribution Image Detection in Neural Networkscited by 1,328
  5. 2017Grad-CAM++: Improved Visual Explanations for Deep Convolutional Networkscited by 939
  6. 2015Towards Open Set Deep Networkscited by 538
  7. 2020Underspecification Presents Challenges for Credibility in Modern Machine Learningcited by 496
  8. 2020CSI: Novelty Detection via Contrastive Learning on Distributionally Shifted Instancescited by 441
  9. 2017Training Confidence-calibrated Classifiers for Detecting Out-of-Distribution Samplescited by 389
  10. 2017The Effectiveness of Data Augmentation in Image Classification using Deep Learningcited by 370
  11. 2021Open-Set Recognition: a Good Closed-Set Classifier is All You Need?cited by 353
  12. 2021Data Augmentation Can Improve Robustnesscited by 194
  13. 2020Contrastive Training for Improved Out-of-Distribution Detectioncited by 186
  14. 2021On the Robustness of Vision Transformers to Adversarial Examplescited by 167
  15. 2015TurkerGaze: Crowdsourcing Saliency with Webcam based Eye Trackingcited by 165
  16. 2017Generative OpenMax for Multi-Class Open Set Classificationcited by 157
  17. 2018Classification-Reconstruction Learning for Open-Set Recognitioncited by 146
  18. 2019C2AE: Class Conditioned Auto-Encoder for Open-set Recognitioncited by 142
  19. 2021OpenGAN: Open-Set Recognition via Open Data Generationcited by 135
  20. 2020Learning Open Set Network with Discriminative Reciprocal Pointscited by 115
  21. 2020Hybrid Models for Open Set Recognitioncited by 109
  22. 2021Pre-training without Natural Imagescited by 100
  23. 2021On Improving Adversarial Transferability of Vision Transformerscited by 70
  24. 2019Data Augmentation Revisited: Rethinking the Distribution Gap between Clean and Augmented Datacited by 33
  25. 2021Can Vision Transformers Learn without Natural Images?cited by 32
  26. 2021Masked Contrastive Learning for Anomaly Detectioncited by 23

Built on by

Cited by year

Citing papers per year: 2022: 2; 2023: 4.

6 papers.

  1. 2022Your Out-of-Distribution Detection Method is Not Robust!cited by 16
  2. 2023A Survey on Open-Set Image Recognitioncited by 7
  3. 2023A Unified Evaluation Framework for Novelty Detection and Accommodation in NLP with an Instantiation in Authorship Attributioncited by 1
  4. 2022Spatial-Temporal Attention Network for Open-Set Fine-Grained Image Recognitioncited by 0
  5. 2023Complementary Frequency-Varying Awareness Network for Open-Set Fine-Grained Image Recognitioncited by 0
  6. 2023Recursive Counterfactual Deconfounding for Object Recognitioncited by 0