AI Paper Atlas

Research map / 3D vision / 3D generation

PerLDiff: Controllable Street View Synthesis Using Perspective-Layout Diffusion Model

Jinhua Zhang, Hualian Sheng, Sijia Cai, Bing Deng, Qiao Liang, Wen Li, Ying Fu, Jieping Ye, Shuhang Gu

ICCV 2025 · 3D vision › 3D generation · approach: Diffusion-based novel view synthesis · 13 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 PerL-based cross-attention for boxes (ablation)BEVControl* · NuScenes valmAP: 16.48 -> 26.07Improved
Add PerL-based cross-attention for road map (ablation)BEVControl* · NuScenes valRoad mIoU: 60.74 -> 61.18Improved
Add both road and box PerL masks (ablation)BEVControl* · NuScenes valmAP: 16.48 -> 25.10Improved
Add both PerL masks (ablation)BEVControl* · NuScenes valFID: 13.05 -> 13.36No clear effect
Add PerLDiff on KITTI trained only on KITTIBEVControl* · KITTI valEasy mAP: 0.33 -> 11.04Improved
Pretrain on NuScenes then fine-tune on KITTIPerLDiff · KITTI valEasy mAP: 11.04 -> 13.12Improved
Increase mask weight lambda from 1.0 to 5.0 (ablation)PerLDiff · NuScenes valmAP: 22.30 -> 25.10Improved
Increase mask weight lambda from 5.0 to 10.0 (ablation)PerLDiff · NuScenes valmAP: 25.10 -> 24.98No clear effect
Replace ConvNext with CLIP road encoder (ablation)PerLDiff · NuScenes valNDS: 36.24 -> 36.26No clear effect
Use GLIGEN-based architecture instead of ControlNet-based (ablation)PerLDiff · NuScenes valmAP: 18.07 -> 25.10Improved
Raise CFG scale from 5.0 to 12.5 (ablation)PerLDiff · NuScenes valmAP: 25.10 -> 23.86Hurt
Add PerL object mask to MagicDriveMagicDrive · NuScenes valNDS: 28.79 -> 29.77Improved

Later papers that compared with this method

1 later accepted paper reports PerLDiff as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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