Method

HaiLong [HL]


Submitted on 30 May. 2016 03:53 by
Four Li (NUST)

Running time:0.16 s
Environment:1 core @ 2.5 Ghz (C/C++)

Method Description:
This model is trained on faster-rcnn.
Parameters:
base_lr: 0.001
lr_policy: "step"
gamma: 0.1
stepsize: 50000
average_loss: 100
momentum: 0.9
weight_decay: 0.0005

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) 77.55 % 64.94 % 50.53 %
Car (Orientation) 41.56 % 35.06 % 27.94 %
Pedestrian (Detection) 58.63 % 42.31 % 34.87 %
Pedestrian (Orientation) 32.32 % 24.21 % 20.43 %
Cyclist (Detection) 55.19 % 39.10 % 32.66 %
Cyclist (Orientation) 30.22 % 21.41 % 17.64 %
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