Impact of realistic properties of the point spread function on classification tasks to reveal a possible distribution shift
Dec 2, 2022
Speakers
Patrick Müller
Speaker · 0 followers
Alexander Braun
Speaker · 0 followers
Margret Keuper
Speaker · 0 followers
About
Image classification is a long-standing task in computer vision with deep neuralnetworks (DNN) producing excellent results on various challenges. However, theyare required not only to perform highly accurate on benchmarks such as ImageNet,but also to robustly handle images in adverse conditions, such as modified lighting, sharpness, weather conditions and image compression. Various benchmarksaimed to measure robustness show that neural networks perform differently wellunder distribution shifts. While datasets such as ImageNet-C model for example common corruptions such as blur and adverse weather conditions, we argue thatthe properties of the optical system and the potentially resulting complex lens blurare insufficiently well studied in the literature. This study evaluates the impact ofrealistic optical corruptions on the ImageNet classification. The proposed complexcorruption kernels are direction and wavelength dependent and include chromaticaberration, which are all to be expected in realistic scenarios such as autonomousdriving applications. Our experiments on twelve different DNN models show significant differences of more than 5Image classification is a long-standing task in computer vision with deep neuralnetworks (DNN) producing excellent results on various challenges. However, theyare required not only to perform highly accurate on benchmarks such as ImageNet,but also to robustly handle images in adverse conditions, such as modified lighting, sharpness, weather conditions and image compression. Various benchmarksaimed to measure robustness show that neural networks perform differently wellunder distribution shifts.…
Organizer
NeurIPS 2022
Account · 952 followers
Like the format? Trust SlidesLive to capture your next event!
Professional recording and live streaming, delivered globally.
Sharing
Recommended Videos
Presentations on similar topic, category or speaker
Could Giant Pre-trained Image Models Extract Universal Representations?
Watch later
Yutong Lin, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Optimistic Posterior Sampling for Reinforcement Learning with Few Samples and Tight Guarantees
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Hyperbolic Deep Reinforcement Learning
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
BiT: Robustly Binarized Multi-distilled Transformer
Watch later
Zechun Liu, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Learning Infinite-Horizon Average-Reward Restless Multi-Action Bandits via Index Awareness
Watch later
Guojun Xiong, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Uncertainty Estimation Using Riemannian Model Dynamics
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%