Semi-supervised Learning for Dense Object Detection in Retail Scenes
Not on the map. Why?
Abstract (arXiv)
Retail scenes usually contain densely packed high number of objects in each image. Standard object detection techniques use fully supervised training methodology. This is highly costly as annotating a large dense retail object detection dataset involves an order of magnitude more effort compared to standard datasets. Hence, we propose semi-supervised learning to effectively use the large amount of unlabeled data available in the retail domain. We adapt a popular self supervised method called noisy student initially proposed for object classification to the task of dense object detection. We show that using unlabeled data with the noisy student training methodology, we can improve the state of the art on precise detection of objects in densely packed retail scenes. We also show that performance of the model increases as you increase the amount of unlabeled data.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 24 of its 32 references that are in the map. First posted on arXiv 5 Jul 2021.
Builds on
24 of 32 references.
References · 24
- 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
- 2014Microsoft COCO: Common Objects in Contextcited by 17,498
- 2018Representation Learning with Contrastive Predictive Codingcited by 7,662
- 2019BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehensioncited by 5,874
- 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,712
- 2013Efficient Estimation of Word Representations in Vector Spacecited by 4,346
- 2020Improved Baselines with Momentum Contrastive Learningcited by 2,282
- 2020FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidencecited by 2,080
- 2018Unsupervised Representation Learning by Predicting Image Rotationscited by 1,586
- 2019MMDetection: Open MMLab Detection Toolbox and Benchmarkcited by 1,379
- 2019Self-training with Noisy Student improves ImageNet classificationcited by 1,304
- 2020CURL: Contrastive Unsupervised Representations for Reinforcement Learningcited by 745
12 more
- 2019Bridging the Gap Between Anchor-based and Anchor-free Detection via Adaptive Training Sample Selectioncited by 539
- 2017Time-Contrastive Networks: Self-Supervised Learning from Videocited by 453
- 2020Learning Invariant Representations for Reinforcement Learning without Reconstructioncited by 424
- 2020Generalized Focal Loss: Learning Qualified and Distributed Bounding Boxes for Dense Object Detectioncited by 408
- 2020A Simple Semi-Supervised Learning Framework for Object Detectioncited by 282
- 2020VarifocalNet: An IoU-aware Dense Object Detectorcited by 230
- 2019A Survey of Deep Learning-based Object Detectioncited by 106
- 2019Region Proposal by Guided Anchoringcited by 87
- 2019Precise Detection in Densely Packed Scenescited by 62
- 2019Contextual Imagined Goals for Self-Supervised Robotic Learningcited by 40
- 2019Benchmark for Generic Product Detection: A Low Data Baseline for Dense Object Detectioncited by 3
- 2020Learning Gaussian Maps for Dense Object Detectioncited by 2
Built on by
Citing papers per year: 2022: 1.
1 paper.