Mini-batch graphs for robust image classification
Not on the map. Why?
Abstract (arXiv)
Current deep learning models for classification tasks in computer vision are trained using mini-batches. In the present article, we take advantage of the relationships between samples in a mini-batch, using graph neural networks to aggregate information from similar images. This helps mitigate the adverse effects of alterations to the input images on classification performance. Diverse experiments on image-based object and scene classification show that this approach not only improves a classifier's performance but also increases its robustness to image perturbations and adversarial attacks. Further, we also show that mini-batch graph neural networks can help to alleviate the problem of mode collapse in Generative Adversarial Networks.
Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 27 of its 35 references that are in the map. First posted on arXiv 22 Apr 2021.
Builds on
27 of 35 references.
References · 27
- 2015Deep Residual Learning for Image Recognitioncited by 51,809
- 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 11,979
- 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,282
- 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,617
- 2017Graph Attention Networkscited by 8,854
- 2016Image-to-Image Translation with Conditional Adversarial Networkscited by 4,510
- 2016Improved Techniques for Training GANscited by 2,938
- 2020Supervised Contrastive Learningcited by 2,803
- 2018Dynamic Graph CNN for Learning on Point Cloudscited by 2,220
- 2016Convolutional Neural Networks on Graphs with Fast Localized Spectral Filteringcited by 2,085
- 2016Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Networkcited by 1,898
- 2017High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANscited by 1,176
15 more
- 2013Spectral Networks and Locally Connected Networks on Graphscited by 1,097
- 2018Representation Learning on Graphs with Jumping Knowledge Networkscited by 875
- 2017Few-Shot Learning with Graph Neural Networkscited by 324
- 2016Learning Convolutional Neural Networks for Graphscited by 314
- 2019Simple Black-box Adversarial Attackscited by 249
- 2015Deep Convolutional Networks on Graph-Structured Datacited by 246
- 2018Generative Adversarial Networks for Extreme Learned Image Compressioncited by 213
- 2018Prior Convictions: Black-Box Adversarial Attacks with Bandits and Priorscited by 145
- 2019Large-scale interactive object segmentation with human annotatorscited by 109
- 2020Embedding Propagation: Smoother Manifold for Few-Shot Classificationcited by 79
- 2018On GANs and GMMscited by 62
- 2020Beyond Homophily in Graph Neural Networks: Current Limitations and Effective Designscited by 25
- 2019Local Label Propagation for Large-Scale Semi-Supervised Learningcited by 5
- 2019Enhancing Generic Segmentation with Learned Region Representationscited by 4
- 2020Affinity Graph Supervision for Visual Recognitioncited by 4
Built on by
Citing papers per year: 2022: 2; 2023: 0; 2024: 1.
3 papers.