Su Y.-C.Chiu T.-H.Chen Y.-Y.Yeh C.-Y.WINSTON HSU2019-07-102019-07-1020139781450324045https://scholars.lib.ntu.edu.tw/handle/123456789/412983https://www.scopus.com/inward/record.uri?eid=2-s2.0-84887457797&doi=10.1145%2f2502081.2502110&partnerID=40&md5=7d649b3086cbe707f1dea338f164c964The rapid development of technologies in both hardware and software have made content-based multimedia services feasi- ble on mobile devices such as smartphones and tablets; and the strong needs for mobile visual search and recognition have been emerging. While many real applications of vi- sual recognition require a large scale recognition systems, the same technologies that support server-based scalable visual recognition may not be feasible on mobile devices due to the resource constraints. Although the client-server framework ensures the scalability, the real-time response subjects to the limitation on network bandwidth. Therefore, the main challenge for mobile visual recognition system should be the recognition bitrate, which is the amount of data transmis- sion under the same recognition performance. For this work, we exploit and compare various strategies such as compact features, feature compression, feature signatures by hash- ing, image scaling, etc., to enable low bitrate mobile visual recognition. We argue that thumbnail image is a competi- Tive candidate for low bitrate visual recognition because it carries multiple features at once and multi-feature fusion is important as the size of semantic space increases. Our eval- uations on two subsets of ImageNet, both contain more than 10,000 images with 19 and 137 categories, verify the efficacy of thumbnail images. We further suggest a new strategy that combines single (local) feature signature and the thumbnail image, which achieves significant bitrate reduction from (av- erage) 102,570 to 4,661 bytes with merely (overall) 10% per- formance degradation. Copyright ? 2013 ACM.Bitrate; Mobile image recognition; Multi- modal fusion; Thumbnail image[SDGs]SDG9Bit rates; Client-server framework; Content-based multimedia; Hardware and software; Mobile visual searches; Multi-feature fusion; Recognition systems; Thumbnail image; Image recognition; Mobile devices; Multimedia services; Semantics; Image codingconference paper10.1145/2502081.25021102-s2.0-84887457797