AI Paper Atlas

Research map / 3D vision / Face and shape

Visual-to-Executable Procedural Reconstruction of Buildings.

Zichong Zhu, Zhihao Liu, Andrei Sharf, Zhanglin Cheng

NeurIPS 2026 · 3D vision › Face and shape · approach: Textured mesh generation · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Studies image-based reconstruction of buildings as executable procedural programs, aiming to replace manually designed grammars with automatically inferred rules.

Problem

Procedural building representations are compact and editable but require manually designed grammars that are hard to obtain in practice; the paper targets recovering them from images.

Approach

Image-based inference of executable procedural rules for buildings, so that reconstructed geometry is encoded as runnable procedural programs rather than static meshes. Details beyond the abstract are not stated.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
predict structured intermediate representation (middle code) instead of direct low-level CGA code (ablation)Gemini 3.1 Pro Preview · Tile-level generationexecution success: 0.68 -> 0.95Improved
use fine-tuned component predictor instead of prompted middle-code generation (ablation)component parameterization · Tile-level generationexecution success: 0.95 -> 0.98Improved
hierarchical pipeline versus direct end-to-end LLM CGA generationfull framework · Full-program comparisonTRC: 1.83 -> 1.47Improved
add first visual-symbolic reflection iteration (ablation)layout induction · Layout recoveryTRC: 4.32 -> 2.01Improved
add second reflection iteration (ablation)layout induction · Layout recoveryTRC: 2.01 -> 1.47Improved
more reflection iterations beyond two (ablation)layout induction · Layout recoverystructural accuracyNo clear effect
combine CLIP and DINOv3 streams versus CLIP-only (ablation)tile encoder · Window tile attributesSill Height accuracy: 0.72 -> 0.92Improved
combine CLIP and DINOv3 streams versus DINOv3-only (ablation)tile encoder · Window tile attributesArch Height accuracy: 0.88 -> 0.90Improved
add real training crops to synthetic training (ablation)component predictor · 107 real held-out tilesoverall macro accuracy: 0.868 -> 0.940Improved

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