Paper Info
Title
End-to-End Compression Towards Machine Vision: Network Architecture Design and Optimization
Abstract
The visual signal compression is a long-standing problem. Fueled by the recent advances of deep learning, exciting progress has been made. Despite better compression performance, existing end-to-end compression algorithms are still designed towards better signal quality in terms of rate-distortion optimization. In this paper, we show that the design and optimization of network architecture could be further improved for compression towards machine vision. We propose an inverted bottleneck structure for the encoder of the end-to-end compression towards machine vision, which specifically accounts for efficient representation of the semantic information. Moreover, we quest the capability of optimization by incorporating the analytics accuracy into the optimization process, and the optimality is further explored with generalized rate-accuracy optimization in an iterative manner. We use object detection as a showcase for end-to-end compression towards machine vision, and extensive experiments show that the proposed scheme achieves significant BD-rate savings in terms of analysis performance. Moreover, the promise of the scheme is also demonstrated with strong generalization capability towards other machine vision tasks, due to the enabling of signal-level reconstruction.
Year
DOI
Venue
2021
10.1109/OJCAS.2021.3126061
IEEE Open Journal of Circuits and Systems
Keywords
DocType
Volume
Visual signal compression,machine vision,object detection,rate-distortion optimization
Journal
2
ISSN
Citations 
PageRank 
2644-1225
0
0.34
References 
Authors
0
4
Name
Order
Citations
PageRank
Shiqi Wang11281120.37
Shiqi Wang21281120.37
Zhao Wang301.01
Yan Ye45412.55