Method

cas_retina_1_13 [cas_retina_1_13]


Submitted on 13 Jan. 2019 04:20 by
Yue Li (Southern Methodist University)

Running time:0.03 s
Environment:4 cores @ 2.5 Ghz (Python)

Method Description:
+0.5 float instead of int
Parameters:
lr=0.0001
Latex Bibtex:

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) 89.22 % 84.43 % 75.39 %
Car (Orientation) 38.39 % 35.68 % 31.89 %
Pedestrian (Detection) 77.11 % 59.87 % 57.81 %
Pedestrian (Orientation) 46.29 % 36.62 % 35.40 %
Cyclist (Detection) 71.59 % 55.81 % 52.16 %
Cyclist (Orientation) 30.31 % 24.37 % 25.81 %
This table as LaTeX


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



Orientation estimation results.
This figure as: png eps pdf txt gnuplot



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



Orientation estimation results.
This figure as: png eps pdf txt gnuplot



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



Orientation estimation results.
This figure as: png eps pdf txt gnuplot




eXTReMe Tracker