Wilds: A benchmark of in-the-wild distribution shifts

PW Koh, S Sagawa, H Marklund… - International …, 2021 - proceedings.mlr.press
Distribution shifts—where the training distribution differs from the test distribution—can
substantially degrade the accuracy of machine learning (ML) systems deployed in the wild …

Benchmarking neural network robustness to common corruptions and perturbations

D Hendrycks, T Dietterich - arXiv preprint arXiv:1903.12261, 2019 - arxiv.org
In this paper we establish rigorous benchmarks for image classifier robustness. Our first
benchmark, ImageNet-C, standardizes and expands the corruption robustness topic, while …

A survey of the four pillars for small object detection: Multiscale representation, contextual information, super-resolution, and region proposal

G Chen, H Wang, K Chen, Z Li, Z Song… - … on systems, man …, 2020 - ieeexplore.ieee.org
Although great progress has been made in generic object detection by advanced deep
learning techniques, detecting small objects from images is still a difficult and challenging …

On robustness and transferability of convolutional neural networks

J Djolonga, J Yung, M Tschannen… - Proceedings of the …, 2021 - openaccess.thecvf.com
Modern deep convolutional networks (CNNs) are often criticized for not generalizing under
distributional shifts. However, several recent breakthroughs in transfer learning suggest that …

Road infrastructure challenges faced by automated driving: A review

T Mihalj, H Li, D Babić, C Lex, M Jeudy, G Zovak… - Applied Sciences, 2022 - mdpi.com
Automated driving can no longer be referred to as hype or science fiction but rather a
technology that has been gradually introduced to the market. The recent activities of …

A guide to image and video based small object detection using deep learning: Case study of maritime surveillance

AM Rekavandi, L Xu, F Boussaid… - arXiv preprint arXiv …, 2022 - arxiv.org
Small object detection (SOD) in optical images and videos is a challenging problem that
even state-of-the-art generic object detection methods fail to accurately localize and identify …

Backpropagated gradient representations for anomaly detection

G Kwon, M Prabhushankar, D Temel… - Computer Vision–ECCV …, 2020 - Springer
Learning representations that clearly distinguish between normal and abnormal data is key
to the success of anomaly detection. Most of existing anomaly detection algorithms use …

Olives dataset: Ophthalmic labels for investigating visual eye semantics

M Prabhushankar, K Kokilepersaud… - Advances in …, 2022 - proceedings.neurips.cc
Clinical diagnosis of the eye is performed over multifarious data modalities including scalar
clinical labels, vectorized biomarkers, two-dimensional fundus images, and three …

Automatic traffic sign detection and recognition using SegU-Net and a modified Tversky loss function with L1-constraint

U Kamal, TI Tonmoy, S Das… - IEEE Transactions on …, 2019 - ieeexplore.ieee.org
Traffic sign detection is a central part of autonomous vehicle technology. Recent advances
in deep learning algorithms have motivated researchers to use neural networks to perform …

Traffic sign detection under challenging conditions: A deeper look into performance variations and spectral characteristics

D Temel, MH Chen, G AlRegib - IEEE Transactions on …, 2019 - ieeexplore.ieee.org
Traffic signs are critical for maintaining the safety and efficiency of our roads. Therefore, we
need to carefully assess the capabilities and limitations of automated traffic sign detection …