Abstract
CNNs have massively improved performance in object detection in photographs. However research into object detection in artwork remains limited. We show state-of-the-art performance on a challenging dataset, People-Art, which contains people from photos, cartoons and 41 different artwork movements. We achieve this high performance by finetuning a CNN for this task, thus also demonstrating that training CNNs on photos results in overfitting for photos: only the first three or four layers transfer from photos to artwork. Although the CNN’s performance is the highest yet, it remains less than 60% AP, suggesting further work is needed for the cross-depiction problem.
| Original language | English |
|---|---|
| Title of host publication | Computer Vision - ECCV 2016 Workshops. ECCV 2016. |
| Editors | Gang Hua, Hervé Jégou |
| Publisher | Springer Verlag |
| Pages | 825-841 |
| Number of pages | 17 |
| ISBN (Print) | 9783319466033 |
| DOIs | |
| Publication status | Published - 18 Sep 2016 |
| Event | 14th European Conference on Computer Vision, ECCV 2016 - Amsterdam, Netherlands Duration: 8 Oct 2016 → 16 Oct 2016 |
Publication series
| Name | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
|---|---|
| Volume | 9913 LNCS |
| ISSN (Print) | 0302-9743 |
| ISSN (Electronic) | 1611-3349 |
Conference
| Conference | 14th European Conference on Computer Vision, ECCV 2016 |
|---|---|
| Country | Netherlands |
| City | Amsterdam |
| Period | 8/10/16 → 16/10/16 |
Keywords
- CNNs
- Cross-depiction problem
- Object recognition
ASJC Scopus subject areas
- Theoretical Computer Science
- Computer Science(all)
Cite this
Detecting people in artwork with CNNs. / Westlake, Nicholas; Cai, Hongping; Hall, Peter.
Computer Vision - ECCV 2016 Workshops. ECCV 2016.. ed. / Gang Hua; Hervé Jégou. Springer Verlag, 2016. p. 825-841 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9913 LNCS).Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Detecting people in artwork with CNNs
AU - Westlake, Nicholas
AU - Cai, Hongping
AU - Hall, Peter
PY - 2016/9/18
Y1 - 2016/9/18
N2 - CNNs have massively improved performance in object detection in photographs. However research into object detection in artwork remains limited. We show state-of-the-art performance on a challenging dataset, People-Art, which contains people from photos, cartoons and 41 different artwork movements. We achieve this high performance by finetuning a CNN for this task, thus also demonstrating that training CNNs on photos results in overfitting for photos: only the first three or four layers transfer from photos to artwork. Although the CNN’s performance is the highest yet, it remains less than 60% AP, suggesting further work is needed for the cross-depiction problem.
AB - CNNs have massively improved performance in object detection in photographs. However research into object detection in artwork remains limited. We show state-of-the-art performance on a challenging dataset, People-Art, which contains people from photos, cartoons and 41 different artwork movements. We achieve this high performance by finetuning a CNN for this task, thus also demonstrating that training CNNs on photos results in overfitting for photos: only the first three or four layers transfer from photos to artwork. Although the CNN’s performance is the highest yet, it remains less than 60% AP, suggesting further work is needed for the cross-depiction problem.
KW - CNNs
KW - Cross-depiction problem
KW - Object recognition
UR - http://www.scopus.com/inward/record.url?scp=84989834205&partnerID=8YFLogxK
U2 - 10.1007/978-3-319-46604-0_57
DO - 10.1007/978-3-319-46604-0_57
M3 - Conference contribution
AN - SCOPUS:84989834205
SN - 9783319466033
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 825
EP - 841
BT - Computer Vision - ECCV 2016 Workshops. ECCV 2016.
A2 - Hua, Gang
A2 - Jégou, Hervé
PB - Springer Verlag
ER -