Just Recognizable Distortion for Machine Vision Oriented Image and Video Coding

Machine visual intelligence has exploded in recent years. Large-scale, high-quality image and video datasets significantly empower learning-based machine vision models, especially deep-learning models. However, images and videos are usually compressed before being analyzed in practical situations wh...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:International journal of computer vision Jg. 129; H. 10; S. 2889 - 2906
Hauptverfasser: Zhang, Qi, Wang, Shanshe, Zhang, Xinfeng, Ma, Siwei, Gao, Wen
Format: Journal Article
Sprache:Englisch
Veröffentlicht: New York Springer US 01.10.2021
Springer
Springer Nature B.V
Schlagworte:
Analysis
Artificial Intelligence
Computer Imaging
Computer Science
Data compression
Datasets
Deep learning
Distortion
Image coding
Image compression
Image Processing and Computer Vision
Image quality
Machine vision
Pattern Recognition
Pattern Recognition and Graphics
Special Issue on Deep Learning for Video Analysis and Compression
Video compression
Vision
Vision systems
Deep learning
Just noticeable distortion
Machine vision
Image and video coding
ISSN:0920-5691, 1573-1405
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Machine visual intelligence has exploded in recent years. Large-scale, high-quality image and video datasets significantly empower learning-based machine vision models, especially deep-learning models. However, images and videos are usually compressed before being analyzed in practical situations where transmission or storage is limited, leading to a noticeable performance loss of vision models. In this work, we broadly investigate the impact on the performance of machine vision from image and video coding. Based on the investigation, we propose Just Recognizable Distortion (JRD) to present the maximum distortion caused by data compression that will reduce the machine vision model performance to an unacceptable level. A large-scale JRD-annotated dataset containing over 340,000 images is built for various machine vision tasks, where the factors for different JRDs are studied. Furthermore, an ensemble-learning-based framework is established to predict the JRDs for diverse vision tasks under few- and non-reference conditions, which consists of multiple binary classifiers to improve the prediction accuracy. Experiments prove the effectiveness of the proposed JRD-guided image and video coding to significantly improve compression and machine vision performance. Applying predicted JRD is able to achieve remarkably better machine vision task accuracy and save a large number of bits.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0920-5691
1573-1405
DOI:10.1007/s11263-021-01505-4