Debiased Novel Category Discovering and Localization (2402.18821v1)
Abstract: In recent years, object detection in deep learning has experienced rapid development. However, most existing object detection models perform well only on closed-set datasets, ignoring a large number of potential objects whose categories are not defined in the training set. These objects are often identified as background or incorrectly classified as pre-defined categories by the detectors. In this paper, we focus on the challenging problem of Novel Class Discovery and Localization (NCDL), aiming to train detectors that can detect the categories present in the training data, while also actively discover, localize, and cluster new categories. We analyze existing NCDL methods and identify the core issue: object detectors tend to be biased towards seen objects, and this leads to the neglect of unseen targets. To address this issue, we first propose an Debiased Region Mining (DRM) approach that combines class-agnostic Region Proposal Network (RPN) and class-aware RPN in a complementary manner. Additionally, we suggest to improve the representation network through semi-supervised contrastive learning by leveraging unlabeled data. Finally, we adopt a simple and efficient mini-batch K-means clustering method for novel class discovery. We conduct extensive experiments on the NCDL benchmark, and the results demonstrate that the proposed DRM approach significantly outperforms previous methods, establishing a new state-of-the-art.
- Measuring the Objectness of Image Windows. IEEE Transactions on Pattern Analysis and Machine Intelligence, 34(11): 2189–2202.
- Towards Open World Recognition. CoRR, abs/1412.5687.
- Unsupervised Object Discovery and Localization in the Wild: Part-based Matching with Bottom-up Region Proposals. CoRR, abs/1501.06170.
- The Overlooked Elephant of Object Detection: Open Set. In 2020 IEEE Winter Conference on Applications of Computer Vision (WACV), 1010–1019.
- Context as Supervisory Signal: Discovering Objects with Predictable Context. In European Conference on Computer Vision (ECCV), 362–377. Springer.
- The pascal visual object classes (voc) challenge. International journal of computer vision, 88(2): 303–338.
- Girshick, R. B. 2015. Fast R-CNN. CoRR, abs/1504.08083.
- Rich feature hierarchies for accurate object detection and semantic segmentation. CoRR, abs/1311.2524.
- Zero-Shot Detection via Vision and Language Knowledge Distillation. CoRR, abs/2104.13921.
- Deep learning with weak annotation from diagnosis reports for detection of multiple head disorders: a prospective, multicentre study. The Lancet Digital Health, 4(8): e584–e593.
- Automatically discovering and learning new visual categories with ranking statistics. arXiv preprint arXiv:2002.05714.
- Autonovel: Automatically discovering and learning novel visual categories. IEEE Transactions on Pattern Analysis and Machine Intelligence, 44(10): 6767–6781.
- Learning to discover novel visual categories via deep transfer clustering. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 8401–8409.
- SECRET: Self-Consistent Pseudo Label Refinement for Unsupervised Domain Adaptive Person Re-identification. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 36, 879–887.
- Learning to cluster in order to transfer across domains and tasks. arXiv preprint arXiv:1711.10125.
- Multi-class open set recognition using probability of inclusion. In European Conference on Computer Vision, 393–409. Springer.
- Towards Open World Object Detection. CoRR, abs/2103.02603.
- Learning Open-World Object Proposals without Learning to Classify. CoRR, abs/2108.06753.
- Microsoft coco: Common objects in context. In European conference on computer vision, 740–755. Springer.
- SSD: Single Shot MultiBox Detector. CoRR, abs/1512.02325.
- The Pursuit of Knowledge: Discovering and Localizing Novel Categories using Dual Memory. CoRR, abs/2105.01652.
- You Only Look Once: Unified, Real-Time Object Detection. CoRR, abs/1506.02640.
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. CoRR, abs/1506.01497.
- Toward open set recognition. IEEE transactions on pattern analysis and machine intelligence, 35(7): 1757–1772.
- Novel visual category discovery with dual ranking statistics and mutual knowledge distillation. Advances in Neural Information Processing Systems, 34: 22982–22994.
- Revisiting Open World Object Detection. CoRR, abs/2201.00471.
- Neighborhood contrastive learning for novel class discovery. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 10867–10875.