Method

TAFT [TAFT]


Submitted on 19 Sep. 2018 09:46 by
Jifeng Shen (Jiangsu university)

Running time:0.2 s
Environment:1 core @ 2.5 Ghz (C/C++)

Method Description:
we use zero-th, first and second order
differential features on the 10-channel feature
maps for pedestrian detection.
Parameters:
OU=2,S=1,R=3,M=9,D=3
Latex Bibtex:
@article{Shen2018 T-ITS,
title = "Differential Features for Pedestrian
Detection: A
Taylor Series Perspective",
journal = "IEEE Transactions on Intelligent
Transportation Systems",

author = "Jifeng Shen and Xin Zuo and Wankou
Yang
and Danil Prokhorov and Xue Mei and Haibin
Ling",
year=2018
}

Detailed Results

Object detection and orientation estimation results. Results for object detection are given in terms of average precision (AP) and results for joint object detection and orientation estimation are provided in terms of average orientation similarity (AOS).


Benchmark Easy Moderate Hard
Pedestrian (Detection) 67.62 % 53.15 % 47.08 %
This table as LaTeX

2D object detection results.
This figure as: png eps pdf txt gnuplot




eXTReMe Tracker