[HTML][HTML] Data augmentation: A comprehensive survey of modern approaches

A Mumuni, F Mumuni - Array, 2022 - Elsevier
To ensure good performance, modern machine learning models typically require large
amounts of quality annotated data. Meanwhile, the data collection and annotation processes …

Image data augmentation approaches: A comprehensive survey and future directions

T Kumar, R Brennan, A Mileo, M Bendechache - IEEE Access, 2024 - ieeexplore.ieee.org
Deep learning algorithms have exhibited impressive performance across various computer
vision tasks; however, the challenge of overfitting persists, especially when dealing with …

When vision transformers outperform resnets without pre-training or strong data augmentations

X Chen, CJ Hsieh, B Gong - arXiv preprint arXiv:2106.01548, 2021 - arxiv.org
Vision Transformers (ViTs) and MLPs signal further efforts on replacing hand-wired features
or inductive biases with general-purpose neural architectures. Existing works empower the …

Enhance the visual representation via discrete adversarial training

X Mao, Y Chen, R Duan, Y Zhu, G Qi… - Advances in …, 2022 - proceedings.neurips.cc
Adversarial Training (AT), which is commonly accepted as one of the most effective
approaches defending against adversarial examples, can largely harm the standard …

[PDF][PDF] Shape-texture debiased neural network training

Y Li, Q Yu, M Tan, J Mei, P Tang, W Shen… - arXiv preprint arXiv …, 2020 - researchgate.net
Shape and texture are two prominent and complementary cues for recognizing objects.
Nonetheless, Convolutional Neural Networks are often biased towards either texture or …

Coco-o: A benchmark for object detectors under natural distribution shifts

X Mao, Y Chen, Y Zhu, D Chen, H Su… - Proceedings of the …, 2023 - openaccess.thecvf.com
Practical object detection application can lose its effectiveness on image inputs with natural
distribution shifts. This problem leads the research community to pay more attention on the …

Adversarial attacks and defenses in deep learning for image recognition: A survey

J Wang, C Wang, Q Lin, C Luo, C Wu, J Li - Neurocomputing, 2022 - Elsevier
In recent years, researches on adversarial attacks and defense mechanisms have obtained
much attention. It's observed that adversarial examples crafted with small malicious …

Gsrformer: Grounded situation recognition transformer with alternate semantic attention refinement

ZQ Cheng, Q Dai, S Li, T Mitamura… - Proceedings of the 30th …, 2022 - dl.acm.org
Grounded Situation Recognition (GSR) aims to generate structured semantic summaries of
images for" human-like''event understanding. Specifically, GSR task not only detects the …

Does robustness on imagenet transfer to downstream tasks?

Y Yamada, M Otani - … of the IEEE/CVF conference on …, 2022 - openaccess.thecvf.com
As clean ImageNet accuracy nears its ceiling, the research community is increasingly more
concerned about robust accuracy under distributional shifts. While a variety of methods have …

Harnessing perceptual adversarial patches for crowd counting

S Liu, J Wang, A Liu, Y Li, Y Gao, X Liu… - Proceedings of the 2022 …, 2022 - dl.acm.org
Crowd counting, which has been widely adopted for estimating the number of people in
safety-critical scenes, is shown to be vulnerable to adversarial examples in the physical …