AI Paper Atlas

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

MonoGRNet: A General Framework for Monocular 3D Object Detection

Zengyi Qin, Jinglu Wang, Yan Lu

TPAMI 2021 · Detection and segmentation › 3D detection and driving · approach: Monocular 3D object detection · 46 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

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
add L_Zc instance depth loss (ablation)MonoGRNet (W) · KITTI val AP3D IoU=0.3 ModerateAP3D: 28.56 -> 32.33Improved
add projected 3D center loss L_c (ablation)MonoGRNet (W) · KITTI val AP3D IoU=0.3 ModerateAP3D: 32.33 -> 40.59Improved
add motion consistency loss L_dC (ablation)MonoGRNet (W) · KITTI val AP3D IoU=0.3 ModerateAP3D: 40.59 -> 41.19Improved
add last loss L_a (ablation)MonoGRNet (W) · KITTI val AP3D IoU=0.3 ModerateAP3D: 41.19 -> 42.61Improved
use geometry-guided learning (GeoGL) instead of MinProjErr for 3D location (ablation)BA x1 orientation · KITTI val AP3D IoU=0.3 ModerateAP3D: 29.80 -> 41.40Improved
use object-centric orientation transfer instead of background-aware x1 (ablation)GeoGL · KITTI val AP3D IoU=0.3 ModerateAP3D: 41.40 -> 42.61Improved
use object-centric orientation transfer instead of background-aware x2 (ablation)GeoGL · KITTI val AP3D IoU=0.3 ModerateAP3D: 39.16 -> 42.61Improved
use object-centric transfer instead of background-aware x1 (ablation)orientation learning · KITTI val AOS IoU=0.3 ModerateAOS: 80.54 -> 88.78Improved
enlarge crop with background (BA x2 vs BA x1) (ablation)orientation learning · KITTI val AOS IoU=0.3 ModerateAOS: 80.54 -> 62.86Hurt
use full 3D supervision (MonoGRNet F) instead of weak 2D supervisionMonoGRNet · KITTI val AP3D IoU=0.5 ModerateAP3D: 21.57 -> 36.20Improved
use LiDAR supervision during training (Ma et al.) vs mono-only MonoGRNet (F)monocular 3D detector · KITTI val AP3D IoU=0.5 ModerateAP3D: 36.20 -> 49.19Improved

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