mlsanityData to 7 Oct 2026

Towards Few-Annotation Learning in Computer Vision: Application to Image Classification and Object Detection tasks

arXiv 2311.04888 ↗First posted 8 Nov 2023cs.CV cs.AI cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

In this thesis, we develop theoretical, algorithmic and experimental contributions for Machine Learning with limited labels, and more specifically for the tasks of Image Classification and Object Detection in Computer Vision. In a first contribution, we are interested in bridging the gap between theory and practice for popular Meta-Learning algorithms used in Few-Shot Classification. We make connections to Multi-Task Representation Learning, which benefits from solid theoretical foundations, to verify the best conditions for a more efficient meta-learning. Then, to leverage unlabeled data when training object detectors based on the Transformer architecture, we propose both an unsupervised pretraining and a semi-supervised learning method in two other separate contributions. For pretraining, we improve Contrastive Learning for object detectors by introducing the localization information. Finally, our semi-supervised method is the first tailored to transformer-based detectors.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 25 of its 280 references that are in the map. First posted on arXiv 8 Nov 2023.

Builds on

25 of 280 references.

References · 25

  1. 2018Representation Learning with Contrastive Predictive Codingcited by 7,662
  2. 2016Layer Normalizationcited by 5,579
  3. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,155
  4. 2015Continuous control with deep reinforcement learningcited by 3,279
  5. 2020Improved Baselines with Momentum Contrastive Learningcited by 2,282
  6. 2020YOLOv4: Optimal Speed and Accuracy of Object Detectioncited by 1,624
  7. 2017Improved Regularization of Convolutional Neural Networks with Cutoutcited by 1,623
  8. 2019Objects as Pointscited by 1,172
  9. 2018On First-Order Meta-Learning Algorithmscited by 1,003
  10. 2014Neural Turing Machinescited by 630
  11. 2017Meta-SGD: Learning to Learn Quickly for Few-Shot Learningcited by 377
  12. 2020A Simple Semi-Supervised Learning Framework for Object Detectioncited by 282
13 more
  1. 2022Estimating the Carbon Footprint of BLOOM, a 176B Parameter Language Modelcited by 264
  2. 2021A Survey on Deep Semi-supervised Learningcited by 250
  3. 2020Few-Shot Learning via Learning the Representation, Provablycited by 213
  4. 2019SimpleShot: Revisiting Nearest-Neighbor Classification for Few-Shot Learningcited by 194
  5. 2020Provable Meta-Learning of Linear Representationscited by 161
  6. 2017Dataset Augmentation in Feature Spacecited by 124
  7. 2018Data Augmentation by Pairing Samples for Images Classificationcited by 96
  8. 2020Are all negatives created equal in contrastive instance discrimination?cited by 50
  9. 2021A Survey of Self-Supervised and Few-Shot Object Detectioncited by 35
  10. 2021Anchor DETR: Query Design for Transformer-Based Object Detectioncited by 30
  11. 2019Revisiting Fine-tuning for Few-shot Learningcited by 29
  12. 2017Extremely Large Minibatch SGD: Training ResNet-50 on ImageNet in 15 Minutescited by 22
  13. 2021How to Train Your MAML to Excel in Few-Shot Classificationcited by 22

Built on by

Not cited yet.