Important note: On September 7, 2026, the KITTI and KITTI-360 servers had to be restored to the backed-up state of August 7, 2026. Results within this period have to be resubmitted.

Method

TRTConv-L [TRTConv-L]


Submitted on 18 Feb. 2026 00:46 by
jisu kang (korea university)

Running time:0.01 s
Environment:1 core @ 2.5 Ghz (Python)

Method Description:
-
Parameters:
-
Latex Bibtex:
@article{KANG2027104985,
title = {Towards real-time 3D object detection in
LiDAR-only scenario: Semi-supervised learning,
point cloud density augmentation and knowledge
distillation},
journal = {Computer Vision and Image
Understanding},
volume = {274},
pages = {104985},
year = {2027},
issn = {1077-3142},
doi =
{https://doi.org/10.1016/j.cviu.2026.104985},
url =
{https://www.sciencedirect.com/science/article/pii
/S1077314226003528},
author = {Jisu Kang and Wooseok Shin and Sung Won
Han},
keywords = {Real-time operation, 3D object
detection, Computer vision, Semi-supervised
learning, Point cloud augmentation, Knowledge
distillation},}

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) 96.58 % 95.73 % 92.97 %
Car (Orientation) 96.37 % 95.41 % 92.57 %
Car (3D Detection) 91.90 % 85.04 % 80.38 %
Car (Bird's Eye View) 95.55 % 92.04 % 87.23 %
This table as LaTeX

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



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



3D object detection results.
This figure as: png eps txt gnuplot



Bird's eye view results.
This figure as: png eps txt gnuplot




eXTReMe Tracker