Details
| Originalsprache | Englisch |
|---|---|
| Titel des Sammelwerks | Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications - 18th Iberoamerican Congress, CIARP 2013, Proceedings |
| Seiten | 456-463 |
| Seitenumfang | 8 |
| Auflage | PART 1 |
| Publikationsstatus | Veröffentlicht - 2013 |
| Veranstaltung | 18th Iberoamerican Congress on Pattern Recognition, CIARP 2013 - Havana, Kuba Dauer: 20 Nov. 2013 → 23 Nov. 2013 |
Publikationsreihe
| Name | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
|---|---|
| Nummer | PART 1 |
| Band | 8258 LNCS |
| ISSN (Print) | 0302-9743 |
| ISSN (elektronisch) | 1611-3349 |
Abstract
Object detection is an important and challenging task in computer vision. In cascaded detectors, a scanned image is passed through a cascade in which all stage detectors have to classify a found object positively. Common detection algorithms use a sliding window approach, resulting in multiple detections of an object. Thus, the merging of multiple detections is a crucial step in post-processing which has a high impact on the final detection performance. First, this paper proposes a novel method formerging multiple detections that exploits intra-cascade confidences using Dempster's Theory of Evidence. The evidence theory allows hereby to model confidence and uncertainty information to compute the overall confidence measure for a detection. Second, this confidence measure is applied to improve the accuracy of the determined object position. The proposed method is evaluated on public object detection benchmarks and is shown to improve the detection performance.
ASJC Scopus Sachgebiete
- Mathematik (insg.)
- Theoretische Informatik
- Informatik (insg.)
Zitieren
- Standard
- Harvard
- Apa
- Vancouver
- BibTex
- RIS
Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications - 18th Iberoamerican Congress, CIARP 2013, Proceedings. PART 1. Aufl. 2013. S. 456-463 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Band 8258 LNCS, Nr. PART 1).
Publikation: Beitrag in Buch/Bericht/Sammelwerk/Konferenzband › Aufsatz in Konferenzband › Forschung › Peer-Review
}
TY - GEN
T1 - Cleaning up multiple detections caused by sliding window based object detectors
AU - Ehlers, Arne
AU - Scheuermann, Björn
AU - Baumann, Florian
AU - Rosenhahn, Bodo
PY - 2013
Y1 - 2013
N2 - Object detection is an important and challenging task in computer vision. In cascaded detectors, a scanned image is passed through a cascade in which all stage detectors have to classify a found object positively. Common detection algorithms use a sliding window approach, resulting in multiple detections of an object. Thus, the merging of multiple detections is a crucial step in post-processing which has a high impact on the final detection performance. First, this paper proposes a novel method formerging multiple detections that exploits intra-cascade confidences using Dempster's Theory of Evidence. The evidence theory allows hereby to model confidence and uncertainty information to compute the overall confidence measure for a detection. Second, this confidence measure is applied to improve the accuracy of the determined object position. The proposed method is evaluated on public object detection benchmarks and is shown to improve the detection performance.
AB - Object detection is an important and challenging task in computer vision. In cascaded detectors, a scanned image is passed through a cascade in which all stage detectors have to classify a found object positively. Common detection algorithms use a sliding window approach, resulting in multiple detections of an object. Thus, the merging of multiple detections is a crucial step in post-processing which has a high impact on the final detection performance. First, this paper proposes a novel method formerging multiple detections that exploits intra-cascade confidences using Dempster's Theory of Evidence. The evidence theory allows hereby to model confidence and uncertainty information to compute the overall confidence measure for a detection. Second, this confidence measure is applied to improve the accuracy of the determined object position. The proposed method is evaluated on public object detection benchmarks and is shown to improve the detection performance.
UR - http://www.scopus.com/inward/record.url?scp=84893176575&partnerID=8YFLogxK
U2 - 10.1007/978-3-642-41822-8_57
DO - 10.1007/978-3-642-41822-8_57
M3 - Conference contribution
AN - SCOPUS:84893176575
SN - 9783642418211
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 456
EP - 463
BT - Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications - 18th Iberoamerican Congress, CIARP 2013, Proceedings
T2 - 18th Iberoamerican Congress on Pattern Recognition, CIARP 2013
Y2 - 20 November 2013 through 23 November 2013
ER -