Improved Object Detection and Pose Using Part-Based Models
(2013) 18th Scandinavian Conference on Image Analysis (SCIA 2013) 7944. p.396-407- Abstract
- Automated object detection is perhaps the most central task of computer vision and arguably the most difficult one. This paper extends previous work on part-based models by using accurate geometric models both in the learning phase and at detection. In the learning phase manual annotations are used to reduce perspective distortion before learning the part-based models. That training is performed on rectified images, leads to models which are more specific, reducing the risk of false positives. At the same time a set of representative object poses are learnt. These are used at detection to remove perspective distortion. The method is evaluated on the bus category of the Pascal dataset with promising results.
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/4249690
- author
- Jiang, Fangyuan LU ; Enqvist, Olof LU ; Kahl, Fredrik LU and Åström, Karl LU
- organization
- publishing date
- 2013
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- host publication
- Lecture Notes in Computer Science (Image Analysis : 18th Scandinavian Conference, SCIA 2013, Espoo, Finland, June 17-20, 2013. Proceedings)
- editor
- Kämäräinen, Joni-Kristian and Koskela, Markus
- volume
- 7944
- pages
- 12 pages
- publisher
- Springer
- conference name
- 18th Scandinavian Conference on Image Analysis (SCIA 2013)
- conference location
- Espoo, Finland
- conference dates
- 2013-06-17 - 2013-06-20
- external identifiers
-
- wos:000342988500038
- scopus:84884494941
- ISSN
- 1611-3349
- 0302-9743
- ISBN
- 978-3-642-38885-9 (print)
- 978-3-642-38886-6 (online)
- DOI
- 10.1007/978-3-642-38886-6_38
- language
- English
- LU publication?
- yes
- id
- 249fef54-fe0e-4c64-8933-4abe413b82a5 (old id 4249690)
- alternative location
- http://link.springer.com/chapter/10.1007/978-3-642-38886-6_38
- date added to LUP
- 2016-04-01 10:56:56
- date last changed
- 2024-08-26 12:05:51
@inproceedings{249fef54-fe0e-4c64-8933-4abe413b82a5, abstract = {{Automated object detection is perhaps the most central task of computer vision and arguably the most difficult one. This paper extends previous work on part-based models by using accurate geometric models both in the learning phase and at detection. In the learning phase manual annotations are used to reduce perspective distortion before learning the part-based models. That training is performed on rectified images, leads to models which are more specific, reducing the risk of false positives. At the same time a set of representative object poses are learnt. These are used at detection to remove perspective distortion. The method is evaluated on the bus category of the Pascal dataset with promising results.}}, author = {{Jiang, Fangyuan and Enqvist, Olof and Kahl, Fredrik and Åström, Karl}}, booktitle = {{Lecture Notes in Computer Science (Image Analysis : 18th Scandinavian Conference, SCIA 2013, Espoo, Finland, June 17-20, 2013. Proceedings)}}, editor = {{Kämäräinen, Joni-Kristian and Koskela, Markus}}, isbn = {{978-3-642-38885-9 (print)}}, issn = {{1611-3349}}, language = {{eng}}, pages = {{396--407}}, publisher = {{Springer}}, title = {{Improved Object Detection and Pose Using Part-Based Models}}, url = {{https://lup.lub.lu.se/search/files/2261756/4730645.pdf}}, doi = {{10.1007/978-3-642-38886-6_38}}, volume = {{7944}}, year = {{2013}}, }