DMCA
Unsupervised incremental learning for improved object detection in a video (2012)
| Venue: | In CVPR |
| Citations: | 9 - 0 self |
BibTeX
@INPROCEEDINGS{Sharma12unsupervisedincremental,
author = {Pramod Sharma and Chang Huang and Ram Nevatia},
title = {Unsupervised incremental learning for improved object detection in a video},
booktitle = {In CVPR},
year = {2012}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Most common approaches for object detection collect thousands of training examples and train a detector in an offline setting, using supervised learning methods, with the objective of obtaining a generalized detector that would give good performance on various test datasets. However, when an offline trained detector is applied on challenging test datasets, it may fail in some cases by not being able to detect some objects or by producing false alarms. We propose an unsupervised multiple instance learning (MIL) based incremental solution to deal with this issue. We introduce an MIL loss function for Real Adaboost and present a tracking based effective unsupervised online sample collection mechanism to collect the online samples for incremental learning. Experiments demonstrate the effectiveness of our approach by improving the performance of a state of the art offline trained detector on the challenging datasets for pedestrian category. 1.
Keyphrases
incremental learning improved object detection good performance test datasets art offline mil loss function incremental solution object detection online sample pedestrian category unsupervised multiple instance learning challenging datasets generalized detector offline setting false alarm real adaboost training example various test datasets common approach
