Method

StereoDistill [StereoDistill]


Submitted on 10 May. 2022 11:32 by
Zhe Liu (Huazhong University of Science and Technology)

Running time:0.4 s
Environment:1 core @ 2.5 Ghz (Python)

Method Description:
In this paper, we propose a cross-modal distillation method named Stereodistil to narrow the gap between the stereo and LiDAR-based approaches via distilling the stereo detectors from the superior LiDAR model at the response level, which is usually overlooked in 3D object detection distillation.
The key designs of Stereodistil are: the X-component Guided Distillation~(XGD) for regression and the Cross-anchor Logit Distillation~(CLD) for classification. In XGD, instead of empirically adopting a threshold to select the high-quality teacher predictions as soft targets, we decompose the predicted 3D box into
sub-components and retain the corresponding part for distillation if the teacher component pilot is consistent with ground truth to largely boost the number of positive predictions and alleviate the mimicking difficulty of the student model. For CLD, we aggregate the probability distribution of all anchors at the same position to encourage the highest probability anchor rather than i
Parameters:
None
Latex Bibtex:
@inproceedings{liu2020tanet,
title={StereoDistill: Pick the Cream from LiDAR for Distilling Stereo-based 3D Object Detection},
author={Liu, Zhe and Ye, Xiaoqing and Tan, Xiao and Ding Errui and Zhou, Yu and Bai, Xiang},
booktitle={Proceedings of the AAAI Conference on Artificial Intelligence},
year={2023}
}

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) 97.61 % 93.43 % 87.71 %
Car (Orientation) 97.57 % 93.29 % 87.48 %
Car (3D Detection) 81.66 % 66.39 % 57.39 %
Car (Bird's Eye View) 89.03 % 78.59 % 69.34 %
Pedestrian (Detection) 69.00 % 55.09 % 50.95 %
Pedestrian (Orientation) 48.49 % 37.58 % 34.41 %
Pedestrian (3D Detection) 44.12 % 32.23 % 28.95 %
Pedestrian (Bird's Eye View) 50.79 % 37.75 % 34.28 %
Cyclist (Detection) 80.92 % 61.46 % 54.64 %
Cyclist (Orientation) 65.65 % 48.99 % 43.14 %
Cyclist (3D Detection) 63.96 % 44.02 % 39.19 %
Cyclist (Bird's Eye View) 69.46 % 48.37 % 42.69 %
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



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



Bird's eye view 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



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



Bird's eye view 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



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



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




eXTReMe Tracker