Understanding How Image Quality Affects Deep Neural Networks
Image quality is an important practical challenge that is often overlooked in the design of machine vision systems. Commonly, machine vision systems are trained and tested on high quality image datasets, yet in practical applications the input images can not be assumed to be of high quality. Recently, deep neural networks have obtained state-of-the-art performance on many machine vision tasks. In this paper we provide an evaluation of 4 state-of-the-art deep neural network models for image classification under quality distortions. We consider five types of quality distortions: blur, noise, contrast, JPEG, and JPEG2000 compression. We show that the existing networks are susceptible to these quality distortions, particularly to blur and noise. These results enable future work in developing deep neural networks that are more invariant to quality distortions.
Code (4)
Tasks
General Classificationimage-classificationImage ClassificationSimilar Papers 제목 키워드 기반
SAR Despeckling using a Denoising Diffusion Probabilistic Model
Speckle is a multiplicative noise which affects all coherent imaging modalities including Synthetic Aperture Radar (SAR) images. The presence of speckle degrades the image quality and adversely affects the performance of…
Change DetectionDenoisingmodelAn Oracle Inequality for Quasi-Bayesian Non-Negative Matrix Factorization
The aim of this paper is to provide some theoretical understanding of quasi-Bayesian aggregation methods non-negative matrix factorization. We derive an oracle inequality for an aggregated estimator. This result holds fo…
Four Things People Should Know About Migraines
Migraine literacy among the public is known to be low, and this lack of understanding has a negative impact on migraineurs' quality of life. To understand this impact, we use text mining methods to study migraine discuss…
"It's trained by non-disabled people": Evaluating How Image Quality Affects Product Captioning with Vision-Language Models
Vision-Language Models (VLMs) are increasingly used by blind and low-vision (BLV) people to identify and understand products in their everyday lives, such as food, personal care items, and household goods. Despite their …
Understanding the Vulnerability of CLIP to Image Compression
CLIP is a widely used foundational vision-language model that is used for zero-shot image recognition and other image-text alignment tasks. We demonstrate that CLIP is vulnerable to change in image quality under compress…
Image CompressionLanguage ModelingLanguage ModellingZero-Shot Learning