CaSP: Class-agnostic Semi-Supervised Pretraining for Detection and Segmentation

• URL: http://arxiv.org/abs/2112.04966v1
• Date: Thu, 9 Dec 2021 14:54:59 GMT
• Title: CaSP: Class-agnostic Semi-Supervised Pretraining for Detection and Segmentation
• Authors: Lu Qi, Jason Kuen, Zhe Lin, Jiuxiang Gu, Fengyun Rao, Dian Li, Weidong Guo, Zhen Wen, Jiaya Jia
• Abstract summary: We propose a novel Class-agnostic Semi-supervised Pretraining (CaSP) framework to achieve a more favorable task-specificity balance. Using 3.6M unlabeled data, we achieve a remarkable performance gain of 4.7% over ImageNet-pretrained baseline on object detection.
• Score: 60.28924281991539
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: To improve instance-level detection/segmentation performance, existing self-supervised and semi-supervised methods extract either very task-unrelated or very task-specific training signals from unlabeled data. We argue that these two approaches, at the two extreme ends of the task-specificity spectrum, are suboptimal for the task performance. Utilizing too little task-specific training signals causes underfitting to the ground-truth labels of downstream tasks, while the opposite causes overfitting to the ground-truth labels. To this end, we propose a novel Class-agnostic Semi-supervised Pretraining (CaSP) framework to achieve a more favorable task-specificity balance in extracting training signals from unlabeled data. Compared to semi-supervised learning, CaSP reduces the task specificity in training signals by ignoring class information in the pseudo labels and having a separate pretraining stage that uses only task-unrelated unlabeled data. On the other hand, CaSP preserves the right amount of task specificity by leveraging box/mask-level pseudo labels. As a result, our pretrained model can better avoid underfitting/overfitting to ground-truth labels when finetuned on the downstream task. Using 3.6M unlabeled data, we achieve a remarkable performance gain of 4.7% over ImageNet-pretrained baseline on object detection. Our pretrained model also demonstrates excellent transferability to other detection and segmentation tasks/frameworks.

Related papers

• Task Specific Pretraining with Noisy Labels for Remote Sensing Image Segmentation [18.598405597933752]
  Self-supervision provides remote sensing a tool to reduce the amount of exact, human-crafted geospatial annotations. In this work, we propose to exploit noisy semantic segmentation maps for model pretraining. The results from two datasets indicate the effectiveness of task-specific supervised pretraining with noisy labels.
  arXiv  Detail & Related papers  (2024-02-25T18:01:42Z)
• Soft Curriculum for Learning Conditional GANs with Noisy-Labeled and Uncurated Unlabeled Data [70.25049762295193]
  We introduce a novel conditional image generation framework that accepts noisy-labeled and uncurated data during training. We propose soft curriculum learning, which assigns instance-wise weights for adversarial training while assigning new labels for unlabeled data. Our experiments show that our approach outperforms existing semi-supervised and label-noise robust methods in terms of both quantitative and qualitative performance.
  arXiv  Detail & Related papers  (2023-07-17T08:31:59Z)
• ProtoCon: Pseudo-label Refinement via Online Clustering and Prototypical Consistency for Efficient Semi-supervised Learning [60.57998388590556]
  ProtoCon is a novel method for confidence-based pseudo-labeling. Online nature of ProtoCon allows it to utilise the label history of the entire dataset in one training cycle. It delivers significant gains and faster convergence over state-of-the-art datasets.
  arXiv  Detail & Related papers  (2023-03-22T23:51:54Z)
• Label-Efficient Object Detection via Region Proposal Network Pre-Training [58.50615557874024]
  We propose a simple pretext task that provides an effective pre-training for the region proposal network (RPN) In comparison with multi-stage detectors without RPN pre-training, our approach is able to consistently improve downstream task performance.
  arXiv  Detail & Related papers  (2022-11-16T16:28:18Z)
• Revisiting Pretraining for Semi-Supervised Learning in the Low-Label Regime [15.863530936691157]
  Semi-supervised learning (SSL) addresses the lack of labeled data by exploiting large unlabeled data through pseudolabeling. Recent studies combined finetuning (FT) from pretrained weights with SSL to mitigate the challenges and claimed superior results in the low-label regime.
  arXiv  Detail & Related papers  (2022-05-06T03:53:25Z)
• Dash: Semi-Supervised Learning with Dynamic Thresholding [72.74339790209531]
  We propose a semi-supervised learning (SSL) approach that uses unlabeled examples to train models. Our proposed approach, Dash, enjoys its adaptivity in terms of unlabeled data selection.
  arXiv  Detail & Related papers  (2021-09-01T23:52:29Z)
• Self-Tuning for Data-Efficient Deep Learning [75.34320911480008]
  Self-Tuning is a novel approach to enable data-efficient deep learning. It unifies the exploration of labeled and unlabeled data and the transfer of a pre-trained model. It outperforms its SSL and TL counterparts on five tasks by sharp margins.
  arXiv  Detail & Related papers  (2021-02-25T14:56:19Z)
• Boosting the Performance of Semi-Supervised Learning with Unsupervised Clustering [10.033658645311188]
  We show that ignoring labels altogether for whole epochs intermittently during training can significantly improve performance in the small sample regime. We demonstrate our method's efficacy in boosting several state-of-the-art SSL algorithms.
  arXiv  Detail & Related papers  (2020-12-01T14:19:14Z)
• Task-Aware Variational Adversarial Active Learning [42.334671410592065]
  We propose task-aware variational adversarial AL (TA-VAAL) that modifies task-agnostic VAAL. Our proposed TA-VAAL outperforms state-of-the-arts on various benchmark datasets for classifications with balanced / imbalanced labels.
  arXiv  Detail & Related papers  (2020-02-11T22:00:48Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.