Exploring Residual and Spatial Consistency for Object Detection

Hao Wang, Ya Zhang, Zhe Xu

2014

Abstract

Local image features show a high degree of repeatability, while their local appearance usually does not bring enough discriminative pattern to obtain a reliable matching. In this paper, we present a new object matching algorithm based on a novel robust estimation of residual consensus and flexible spatial consistency filter. We evaluate the similarity between different homography model via two-parameter integrated Weibull distribution and inlier probabilities estimates, which can select uncontaminated model to help eliminating outliers. Spatial consistency test was encoded by the geometric relationships of domain knowledge in two directions, which is invariant to scale, rotation, and translation especially robust to the flipped image. Experiment results on nature images with clutter background demonstrate our method effectiveness and robustness.

Paper Citation

in Harvard Style

Wang H., Zhang Y. and Xu Z. (2014). Exploring Residual and Spatial Consistency for Object Detection . In Proceedings of the 9th International Conference on Computer Vision Theory and Applications - Volume 2: VISAPP, (VISIGRAPP 2014) ISBN 978-989-758-004-8, pages 191-197. DOI: 10.5220/0004746801910197

in Bibtex Style

@conference{visapp14,
author={Hao Wang and Ya Zhang and Zhe Xu},
title={Exploring Residual and Spatial Consistency for Object Detection},
booktitle={Proceedings of the 9th International Conference on Computer Vision Theory and Applications - Volume 2: VISAPP, (VISIGRAPP 2014)},
year={2014},
pages={191-197},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0004746801910197},
isbn={978-989-758-004-8},
}

in EndNote Style

TY - CONF
JO - Proceedings of the 9th International Conference on Computer Vision Theory and Applications - Volume 2: VISAPP, (VISIGRAPP 2014)
TI - Exploring Residual and Spatial Consistency for Object Detection
SN - 978-989-758-004-8
AU - Wang H.
AU - Zhang Y.
AU - Xu Z.
PY - 2014
SP - 191
EP - 197
DO - 10.5220/0004746801910197