AI Paper Atlas

Research map / Point clouds / Segmentation and detection

Weakly Supervised Monocular 3D Object Detection Using Multi-View Projection and Direction Consistency

Runzhou Tao, Wencheng Han, Zhongying Qiu, Cheng-Zhong Xu, Jianbing Shen

CVPR 2023 · Point clouds › Segmentation and detection · approach: LiDAR voxel and pillar detectors · 26 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 multi-view consistency loss and direction loss on top of projection loss (ablation)DD3D with WeakMono3D · KITTI val carAP_3D Easy: 0.00 -> 49.37 (full)Improved
remove direction consistency loss (keep projection + multi-view) (ablation)WeakMono3D · KITTI val carAP_3D Easy: 49.37 -> 11.78 when removedHurt
add multi-view consistency loss to projection loss (ablation)WeakMono3D · KITTI val carAP_3D Easy: 0.00 -> 11.78Improved
use multi-camera pairs instead of single-camera video sequences (ablation)WeakMono3D · KITTI val carAP_3D Easy: 36.68 -> 49.37Improved
pre-train with WeakMono3D then fine-tune with 1/3 3D labels (ablation)DD3D · KITTI val carAP_3D Easy: 58.08 -> 61.02Improved
pre-train with WeakMono3D then fine-tune with varying small label ratios (1/20 to 1/5) (ablation)DD3D · KITTI val carAP_3D Easy: 1/20: 53.61; 1/10: 57.21; 1/5: 59.12 vs DD3D 58.08Improved
weakly supervised 2D-label training vs WeakM3D (LiDAR-based weak)WeakMono3D · KITTI test car IoU 0.7AP_3D Easy: 5.03 -> 6.98Improved
weakly supervised 2D-label training on pedestrians/cyclists vs fully supervised DD3DWeakMono3D · KITTI test pedestrian/cyclistAP_3D Easy: pedestrian 13.91 (DD3D) vs 4.62 (ours)Hurt

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