Memory-efficient image databases for mobile visual search
Abstract/Contents
- Abstract
- Mobile visual search (MVS) systems compare query images captured by the mobile device's camera against a database of labeled images to recognize objects seen in the device's viewfinder. Practical MVS systems require a fast response to provide an interactive and compelling user experience. We show how a memory-efficient database stored on a mobile device can effectively enhance the capabilities of MVS systems and enable fast and accurate queries in different environments. The image signatures stored in the database on the mobile device must be compact to fit in the device's small memory capacity, capable of fast comparisons across a large database, and robust against large visual distortions. We first develop two methods for efficiently compressing a database constructed from feature histograms. Our methods reduce the database memory usage by 4-5 times without any loss in matching accuracy and possess fast decoding capabilities. Subsequently, we develop a third database representation based on feature residuals that is even more compact. Compressed residuals reduce the database memory usage by 12-14 times, require only a small codebook, and enable image matching directly in the compressed domain. With a compact database stored on a mobile device, we implement an MVS system that can recognize media covers, book spines, outdoor landmarks, artwork, and video frames in less than 1 second per query. Our system uses robust motion analysis on the device to automatically infer user interest, select high-quality query frames, and update the pose of recognized objects for accurate augmentation. We also demonstrate how a continuous stream of compact image signatures enables a low bitrate query expansion onto a remote server. The query expansion improves image matching during the current query and updates the local on-device database to benefit future queries.
Description
Type of resource | text |
---|---|
Form | electronic; electronic resource; remote |
Extent | 1 online resource. |
Publication date | 2014 |
Issuance | monographic |
Language | English |
Creators/Contributors
Associated with | Chen, David M |
---|---|
Associated with | Stanford University, Department of Electrical Engineering. |
Primary advisor | Girod, Bernd |
Thesis advisor | Girod, Bernd |
Thesis advisor | Gray, Robert |
Thesis advisor | Grzeszczuk, Radek, 1967- |
Advisor | Gray, Robert |
Advisor | Grzeszczuk, Radek, 1967- |
Subjects
Genre | Theses |
---|
Bibliographic information
Statement of responsibility | David M. Chen. |
---|---|
Note | Submitted to the Department of Electrical Engineering. |
Thesis | Thesis (Ph.D.)--Stanford University, 2014. |
Location | electronic resource |
Access conditions
- Copyright
- © 2014 by David Mo Chen
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...