We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations

DBLP - CS Bibliography


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Computer Vision and Pattern Recognition

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.
Subjects: Computer Vision and Pattern Recognition (cs.CV); Multimedia (cs.MM); Image and Video Processing (eess.IV)
Cite as: arXiv:2107.00328 [cs.CV]
  (or arXiv:2107.00328v2 [cs.CV] for this version)

Submission history

From: Shurun Wang [view email]
[v1] Thu, 1 Jul 2021 09:36:32 GMT (7556kb,D)
[v2] Fri, 19 Nov 2021 04:06:16 GMT (7587kb,D)

Link back to: arXiv, form interface, contact.