DMCA
Robust multi-resolution pedestrian detection in traffic scenes (2013)
| Venue: | In CVPR |
| Citations: | 15 - 2 self |
BibTeX
@INPROCEEDINGS{Yan13robustmulti-resolution,
author = {Junjie Yan and Xucong Zhang and Zhen Lei and Shengcai Liao and Stan Z. Li},
title = {Robust multi-resolution pedestrian detection in traffic scenes},
booktitle = {In CVPR},
year = {2013}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
The serious performance decline with decreasing resolu-tion is the major bottleneck for current pedestrian detection techniques [14, 23]. In this paper, we take pedestrian de-tection in different resolutions as different but related prob-lems, and propose a Multi-Task model to jointly consider their commonness and differences. The model contains res-olution aware transformations to map pedestrians in differ-ent resolutions to a common space, where a shared detector is constructed to distinguish pedestrians from background. For model learning, we present a coordinate descent proce-dure to learn the resolution aware transformations and de-formable part model (DPM) based detector iteratively. In traffic scenes, there are many false positives located around vehicles, therefore, we further build a context model to sup-press them according to the pedestrian-vehicle relationship. The context model can be learned automatically even when the vehicle annotations are not available. Our method re-duces the mean miss rate to 60 % for pedestrians taller than 30 pixels on the Caltech Pedestrian Benchmark, which no-ticeably outperforms previous state-of-the-art (71%). 1.
Keyphrases
traffic scene robust multi-resolution pedestrian detection context model pedestrian-vehicle relationship previous state-of-the-art res-olution aware transformation method re-duces resolution aware transformation current pedestrian detection technique many false positive caltech pedestrian benchmark coordinate descent proce-dure mean miss rate de-formable part model multi-task model major bottleneck vehicle annotation different resolution model learning related prob-lems differ-ent resolution serious performance decline pedestrian de-tection common space
