ThumbNet: One Thumbnail Image Contains All You Need for Recognition
Type
Conference PaperAuthors
Zhao, ChenGhanem, Bernard

KAUST Department
Visual Computing Center (VCC)Electrical Engineering Program
Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
Physical Science and Engineering (PSE) Division
KAUST Grant Number
OSR-CRG2017-3405Date
2020-10-12Permanent link to this record
http://hdl.handle.net/10754/660666
Metadata
Show full item recordAbstract
Although deep convolutional neural networks (CNNs) have achieved great success in computer vision tasks, its real-world application is still impeded by its voracious demand of computational resources. Current works mostly seek to compress the network by reducing its parameters or parameter-incurred computation, neglecting the influence of the input image on the system complexity. Based on the fact that input images of a CNN contain substantial redundancy, in this paper, we propose a unified framework, dubbed as ThumbNet, to simultaneously accelerate and compress CNN models by enabling them to infer on one thumbnail image. We provide three effective strategies to train ThumbNet. In doing so, ThumbNet learns an inference network that performs equally well on small images as the original-input network on large images. With ThumbNet, not only do we obtain the thumbnail-input inference network that can drastically reduce computation and memory requirements, but also we obtain an image downscaler that can generate thumbnail images for generic classification tasks. Extensive experiments show the effectiveness of ThumbNet, and demonstrate that the thumbnail-input inference network learned by ThumbNet can adequately retain the accuracy of the original-input network even when the input images are downscaled 16 times.Citation
Zhao, C., & Ghanem, B. (2020). ThumbNet: One Thumbnail Image Contains All You Need for Recognition. Proceedings of the 28th ACM International Conference on Multimedia. doi:10.1145/3394171.3413937Sponsors
This publication is based upon work supported by the King Abdullah University of Science and Technology (KAUST) Office of Sponsored Research (OSR) under Award No. OSR-CRG2017-3405.Publisher
ACMConference/Event name
28th ACM International Conference on MultimediaISBN
9781450379885arXiv
1904.05034Additional Links
https://dl.acm.org/doi/10.1145/3394171.3413937ae974a485f413a2113503eed53cd6c53
10.1145/3394171.3413937
Scopus Count
Except where otherwise noted, this item's license is described as Archived with thanks to ACM. This work is licensed under a Creative Commons Attribution International 4.0 License.