Please use this identifier to cite or link to this item:
https://hdl.handle.net/2440/108053
Citations | ||
Scopus | Web of Science® | Altmetric |
---|---|---|
?
|
?
|
Type: | Conference paper |
Title: | Beyond Photo-Domain Object Recognition: Benchmarks for the Cross-Depiction Problem |
Author: | Cai, H. Wu, Q. Hall, P. |
Citation: | Proceedings / IEEE International Conference on Computer Vision. IEEE International Conference on Computer Vision, 2015, vol.2015-February, pp.74-79 |
Publisher: | IEEE |
Issue Date: | 2015 |
ISBN: | 9781467383905 |
ISSN: | 1550-5499 |
Conference Name: | IEEE International Conference on Computer Vision Workshops (ICCVW) (11 Dec 2015 - 18 Dec 2015 : Santigo) |
Statement of Responsibility: | Hongping Cai, Qi Wu, Peter Hall |
Abstract: | The cross-depiction problem is that of recognising visual objects regardless of whether they are photographed, painted, drawn, etc. It introduces great challenge as the variance across photo and art domains is much larger than either alone. We extensively evaluate classification, domain adaptation and detection benchmarks for leading techniques, demonstrating that none perform consistently well given the cross-depiction problem. Finally we refine the DPM model, based on query expansion, enabling it to bridge the gap across depiction boundaries to some extent. |
Rights: | © 2015 IEEE |
DOI: | 10.1109/ICCVW.2015.19 |
Published version: | http://dx.doi.org/10.1109/iccvw.2015.19 |
Appears in Collections: | Aurora harvest 8 Computer Science publications |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
RA_hdl_108053.pdf Restricted Access | Restricted Access | 400.06 kB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.