AI Paper Atlas

Research map / 3D vision / 3D generation

RoScenes: A Large-scale Multi-view 3D Dataset for Roadside Perception

Xiaosu Zhu, Hualian Sheng, Sijia Cai, Bing Deng, Shaopeng Yang, Qiao Liang, Ken Chen, Lianli Gao, Jingkuan Song, Jieping Ye

ECCV 2024 · 3D vision › 3D generation · approach: 3D perception for autonomous driving · 12 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
use feature-guided position embedding (RoBEV) instead of static position embedding (PETRv2)RoBEV · RoScenes val avgNDS: 0.627 -> 0.713Improved
use learnable shared position encoding instead of sinusoidal (ablation)RoBEV · RoScenes val avgNDS: 0.608 -> 0.713Improved
share position encoding network between queries and points instead of separate networks (ablation)RoBEV · RoScenes val avgNDS: 0.685 -> 0.713Improved
use implicit BEV methods instead of explicit BEV methodsBEV detectors · RoScenes valNDS: 19.7% higher on averageImproved
use explicit BEV methods (training time and memory)BEV detectors · RoScenestraining time and inference memory: 1.75x training time, 1.50x inference memoryHurt
use nuScenes-pretrained VoVNet instead of ImageNet-pretrained (ablation)RoBEV · RoScenes valNDS: 0.6% drop for ImageNet pretrainingImproved
scale backbone to Swin-B, ViT-L or InternImage-XL (ablation)RoBEV · RoScenes val avgNDS: 0.713 -> 0.719No clear effect
train on all scenes instead of a single scene and test on all scenes (ablation)RoBEV · RoScenes val easyNDS: 0.396 -> 0.753Improved
apply image registration to perturbed images (ablation)RoBEV · pRoScenes easyNDS: 0.441 -> 0.745Improved
use RoBEV on single-layout nuScenesRoBEV · nuScenesNDS: 0.503 -> 0.505No clear effect
increase detection queries from 450 to 900 (ablation)PETRv2 and RoBEV · RoScenes valNDS: 7.7% average improvementImproved
evaluate trained BEV methods on unseen scenesall BEV methods · RoScenes zero-shot testNDS: 0.713 -> 0.308 (RoBEV)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