Research map / Adversarial robustness and security / Attacks on text and graphs
Towards Viewpoint-Invariant Visual Recognition via Adversarial Training
Shouwei Ruan, Yinpeng Dong, Hang Su, Jianteng Peng, Ning Chen, Xingxing Wei
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| apply VIAT (adversarial viewpoint training) to a standard-trained classifier | ResNet-50 · IM3D, GMVFool viewpoints | top-1 accuracy (%): 8.98 -> 59.61 | Improved |
| apply VIAT to a standard-trained classifier | ViT-B/16 · IM3D, GMVFool viewpoints | top-1 accuracy (%): 29.10 -> 83.13 | Improved |
| replace ViewFool with GMVFool (Gaussian mixture) as inner maximization (ablation) | VIAT ResNet-50 · IM3D, ViewFool viewpoints | top-1 accuracy (%): 55.12 -> 59.84 | Improved |
| replace data augmentation with random viewpoints by VIAT (ablation) | ResNet-50 · IM3D, GMVFool viewpoints | top-1 accuracy (%): 33.52 -> 59.61 | Improved |
| apply VIAT on clean ImageNet subset accuracy | ViT-B/16 · ImageNet subset val | top-1 accuracy (%): 92.88 -> 92.56 | No clear effect |
| increase number of Gaussian components K in GMVFool (ablation) | VIAT · IM3D | robust accuracy: K=1 entropy -10.17 vs K=5 -0.69; attack rate 92.13 vs 91.56 | Mixed |
| add distribution sharing across objects in same class (pi=0.5) (ablation) | VIAT · IM3D | viewpoint robustness | Improved |
| evaluate VIAT-trained ResNet-50 on out-of-distribution datasets | ResNet-50 · ImageNet-V | accuracy (%): 28.15 -> 38.96 | Improved |
| evaluate VIAT-trained ResNet-50 on ImageNet-A | ResNet-50 · ImageNet-A | accuracy (%): 19.03 -> 20.32 | Improved |
| use data augmentation with random viewpoint renderings on real OOD datasets (ablation) | ResNet-50 · ImageNet-A | accuracy (%): 19.03 -> 18.65 | No clear effect |
Closest papers by meaning
- Improving Viewpoint Robustness for Visual Recognition via Adversarial Training (TPAMI 2026 · similarity 0.98)
- ViewFool: Evaluating the Robustness of Visual Recognition to Adversarial Viewpoints (NeurIPS 2022 · similarity 0.91)
- Omniview-Tuning: Boosting Viewpoint Invariance of Vision-Language Pre-training Models (ECCV 2024 · similarity 0.86)
- Vision Transformers Beat WideResNets on Small Scale Datasets Adversarial Robustness (AAAI 2025 · similarity 0.84)
- Gain from Neighbors: Boosting Model Robustness in the Wild via Adversarial Perturbations Toward Neighboring Classes (CVPR 2025 · similarity 0.84)
- Rethinking Invariance Regularization in Adversarial Training to Improve Robustness-Accuracy Trade-off (ICLR 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
