Method

CL-attention [CLA]
[Anonymous Submission]

Submitted on 12 Dec. 2018 19:18 by
[Anonymous Submission]

Running time:0.3 s
Environment:GPU @ 2.5 Ghz (Matlab + C/C++)

Method Description:
Location-aware deformable + attention
Parameters:
Learning rate=0.0005
Latex Bibtex:
@InProceedings{Zhang_2019_CVPR,
author = {Zhang, Chen and Kim, Joohee},
title = {Object Detection With Location-Aware
Deformable Convolution and Backward Attention
Filtering},
booktitle = {The IEEE Conference on Computer
Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2019}
}

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
Car (Detection) 90.51 % 88.99 % 75.50 %
Pedestrian (Detection) 84.26 % 74.03 % 68.45 %
Cyclist (Detection) 82.42 % 74.68 % 65.11 %
This table as LaTeX


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



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



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




eXTReMe Tracker