Improving Robustness for Pose Estimation via Stable Heatmap Regression
Not on the map. Why?
Abstract (arXiv)
Deep learning methods have achieved excellent performance in pose estimation, but the lack of robustness causes the keypoints to change drastically between similar images. In view of this problem, a stable heatmap regression method is proposed to alleviate network vulnerability to small perturbations. We utilize the correlation between different rows and columns in a heatmap to alleviate the multi-peaks problem, and design a highly differentiated heatmap regression to make a keypoint discriminative from surrounding points. A maximum stability training loss is used to simplify the optimization difficulty when minimizing the prediction gap of two similar images. The proposed method achieves a significant advance in robustness over state-of-the-art approaches on two benchmark datasets and maintains high performance.
Key facts. Cited by 1 paper in the map (data through 13 Oct 2026). Builds on 30 of its 41 references that are in the map. First posted on arXiv 8 May 2021.
Builds on
30 of 41 references.
References · 30
- 2015Deep Residual Learning for Image Recognitioncited by 51,950
- 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,321
- 2014Explaining and Harnessing Adversarial Examplescited by 7,336
- 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,379
- 2017mixup: Beyond Empirical Risk Minimizationcited by 5,170
- 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,786
- 2019CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Featurescited by 2,882
- 2019Benchmarking Neural Network Robustness to Common Corruptions and Perturbationscited by 2,862
- 2019Theoretically Principled Trade-off between Robustness and Accuracycited by 1,511
- 2019Deep High-Resolution Representation Learning for Visual Recognitioncited by 1,418
- 2018ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Designcited by 1,134
- 2016Stacked Hourglass Networks for Human Pose Estimationcited by 913
18 more
- 2019AugMix: A Simple Data Processing Method to Improve Robustness and Uncertaintycited by 899
- 2018Simple Baselines for Human Pose Estimation and Trackingcited by 502
- 2019Unlabeled Data Improves Adversarial Robustnesscited by 416
- 2019A Fourier Perspective on Model Robustness in Computer Visioncited by 288
- 2017Learning to Estimate 3D Hand Pose from Single RGB Imagescited by 234
- 2018Adversarial Logit Pairingcited by 224
- 2018Why do deep convolutional networks generalize so poorly to small image transformations?cited by 198
- 2018Lipschitz-Margin Training: Scalable Certification of Perturbation Invariance for Deep Neural Networkscited by 194
- 2019Adversarial Training and Robustness for Multiple Perturbationscited by 183
- 2019Distribution-Aware Coordinate Representation for Human Pose Estimationcited by 165
- 2016Improving the Robustness of Deep Neural Networks via Stability Trainingcited by 156
- 2017GANerated Hands for Real-time 3D Hand Tracking from Monocular RGBcited by 147
- 2019Improving Robustness Without Sacrificing Accuracy with Patch Gaussian Augmentationcited by 105
- 2019Enhancing Adversarial Defense by k-Winners-Take-Allcited by 51
- 2018Towards Highly Accurate and Stable Face Alignment for High-Resolution Videoscited by 19
- 2019Achieving Generalizable Robustness of Deep Neural Networks by Stability Trainingcited by 7
- 2018Towards Robust Neural Networks with Lipschitz Continuitycited by 6
- 2016On The Stability of Video Detection and Trackingcited by 4
Built on by
Citing papers per year: 2023: 1.
1 paper.