AI Paper Atlas

Research map / Detection and segmentation / 3D detection and driving

Rethinking Pseudo-LiDAR Representation

Xinzhu Ma, Shinan Liu, Zhiyi Xia, Hongwen Zhang, Xingyu Zeng, Wanli Ouyang

ECCV 2020 · Detection and segmentation › 3D detection and driving · approach: Depth-based and label-free 3D detection · 214 citations (OpenAlex)

Read the paper

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
represent generated 3D data as image patches processed by 2D CNN instead of point cloud with PointNetpseudo-LiDAR · KITTI val Car monocularAP_3D R11 Moderate: 18.4 -> 18.4No clear effect
use world coordinates (x,y) instead of image coordinates (u,v) (ablation)PatchNet · KITTI val Car monocularAP_3D R11 Moderate: 15.7 -> 22.0Improved
add the y coordinate to the {x,z} input (ablation)PatchNet · KITTI val Car monocularAP_3D R11 Moderate: 18.3 -> 22.0Improved
add x and y coordinates to depth-only input (ablation)PatchNet · KITTI val Car monocularAP_3D R11 Moderate: 3.48 -> 22.0Improved
use 2D CNN backbone ResNet-18 instead of PointNet-18 (ablation)PatchNet-vanilla · KITTI val Car monocularAP_3D R11 Hard: 17.0 -> 19.1Improved
use SE-ResNet-18 instead of ResNet-18 (ablation)PatchNet · KITTI val Car monocularAP_3D R11 Moderate: 21.3 -> 21.5No clear effect
use ResNeXt-101 instead of ResNeXt-50 (ablation)PatchNet · KITTI val Car monocularAP_3D R11 Hard: 17.3 -> 17.0Hurt
add mask global pooling (max) instead of standard global max pooling (ablation)PatchNet · KITTI val Car monocularAP_3D R11 Easy: 32.4 -> 35.1Improved
use mask max pooling instead of mask average pooling (ablation)PatchNet · KITTI val Car monocularAP_3D R11 Moderate: 21.6 -> 22.0Improved
add instance assignment with difficulty-specific heads (distance switcher) (ablation)PatchNet · KITTI val Car monocularAP_3D R11 Easy: 33.7 -> 35.1Improved
use distance instead of predicted difficulty as switcher (ablation)PatchNet · KITTI val Car monocularAP_3D R11 Moderate: 22.1 -> 22.0No clear effect
use PatchNet instead of AM3D on test set Easymonocular 3D detection · KITTI test CarAP_3D R40 Easy: 16.50 -> 15.68Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy