| A Real-Time Deformable Detector. | |
| | |
MedLine Citation:
|
PMID: 21670479 Owner: NLM Status: Publisher |
Abstract/OtherAbstract:
|
We propose a new learning strategy for object detection. The proposed scheme forgoes the need to train a collection of detectors dedicated to homogeneous families of poses, and instead learns a single classifier that has the inherent ability to deform based on the signal of interest.We train a detector with a standard AdaBoost procedure by using combinations of pose-indexed features and pose estimators. This allows the learning process to select and combine various estimates of the pose with features able to compensate for variations in pose without the need to label data for training or explore the pose space in testing. We validate our framework on three types of data: hand video sequences, aerial images of cars as well as face images. We compare our method to a standard boosting framework, with access to the same ground truth, and show a reduction in the false alarm rate of up to an order of magnitude. Where possible, we compare our method to the state-of-the art, which requires pose annotations of the training data, and demonstrate comparable performance. |
| | |
Authors:
|
Karim Ali; François Fleuret; David Hasler; Pascal Fua |
Related Documents
:
|
10959019 - A comparison of manikin cpr performance by lay persons trained in three variations of b... 17488269 - The automated external defibrillator. 9580719 - The preparedness of students to discuss end-of-life issues with patients. 21062569 - Resuscitation by video in northern communities. 20735249 - Astrobiological benefits of human space exploration. 12009219 - Towards new models of cardiopulmonary resuscitation teaching: the role of practical sce... 22464509 - Factors affecting career preferences of medical students at the college of medicine, ma... 19047879 - Beta testing of a network-based health literacy program tailored for older adults with ... 21330799 - The effect of a low-fidelity model on cystoscopic skill training: a single-blinded rand... |
Publication Detail:
|
Type: JOURNAL ARTICLE Date: 2011-6-6 |
Journal Detail:
|
Title: IEEE transactions on pattern analysis and machine intelligence Volume: - ISSN: 1939-3539 ISO Abbreviation: - Publication Date: 2011 Jun |
Date Detail:
|
Created Date: 2011-6-14 Completed Date: - Revised Date: - |
Medline Journal Info:
|
Nlm Unique ID: 9885960 Medline TA: IEEE Trans Pattern Anal Mach Intell Country: - |
Other Details:
|
Languages: ENG Pagination: - Citation Subset: - |
Affiliation:
|
École Polytechnique Fédérale de Lausanne (EPFL), Lausanne and Electronic Microtechnology (CSEM), Neuchâtel. |
Export Citation:
|
APA/MLA Format Download EndNote Download BibTex |
| MeSH Terms | |
Descriptor/Qualifier:
|
|
From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine
Previous Document: Tumor Satellite in Predicting Occult Nodal Metastasis of Tongue Cancer.
Next Document: Object-Graphs for Context-Aware Visual Category Discovery.