Method

Instance Centric Densification via Panoptic Segmentation and Object Completion [ICD-PSOC]


Submitted on 6 Nov. 2025 17:31 by
Yahya Tauk (Université de Technologie de Compiègne)

Running time:0.1 s
Environment:1 core @ 2.5 Ghz (C/C++)

Method Description:
We developped an object detection amthod that uses
object completion focused with panoptic segmentation
to get a more efficient densification of ooi .
Parameters:
TBD
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) 96.19 % 97.83 % 93.03 %
Car (Orientation) 96.15 % 97.63 % 92.77 %
Car (3D Detection) 91.66 % 85.07 % 80.08 %
Car (Bird's Eye View) 95.27 % 91.76 % 88.49 %
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