Awesome AI4CAD Hub Papers · Datasets · Projects
← Back to papers

SOV-CAD: Stepwise Orthographic Views Guided CAD Modeling Sequence Reconstruction

arXiv 2026 66.1 method

TLDR

SOV-CAD uses stepwise orthographic views and offline reinforcement learning to reconstruct CAD modeling sequences from images, outperforming prior methods.

Reasoning

The paper introduces a novel approach combining visual feedback from orthographic projections with a Decision Transformer for sequential CAD reconstruction, showing strong empirical results and data efficiency. However, the abstract lacks explicit details on real-world benchmarks or datasets, though extensive experiments are mentioned.

Read-first score

Read-first score 66.1, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 77.

Recency 8%
100

Uses a gentle age decay so recent papers surface without erasing older foundations. 2026

Reproducibility 25%
81

Screens links and visible text for paper, code, dataset, artifact, and repository signals. pdf=True; code=True; dataset=False; markers=code,github

Methodology quality 25%
70

Screens visible abstract and analysis fields for experiment, dataset, baseline, metric, and limitation evidence. markers=experiment,metric,result

Topical relevance 42%
48.1

Uses existing LLM keyword relevance scores normalized to 0-100. AI for CAD,computer-aided design,neural CAD,generative CAD,parametric CAD,B-Rep,boundary representation,constructive solid geometry,CSG,sketch extrusion,CAD generation,CAD reconstruction,text-to-CAD,image-to-CAD,point cloud to CAD,CAD program

Field roles

FrontierMethodology anchorReproducibility anchor

Rank sensitivity

Stability: volatile; rank range: 17.

Keyword Scores

computer-aided design
10
CAD reconstruction
10
image-to-CAD
9
AI for CAD
8
CAD program
8
neural CAD
7
parametric CAD
7
generative CAD
6
sketch extrusion
6
CAD generation
6
B-Rep
0
boundary representation
0
constructive solid geometry
0
CSG
0
text-to-CAD
0
point cloud to CAD
0

Deep Analysis

Innovations

  • Stepwise visual supervision with orthographic projections at each modeling step (target, current model, active sketch) for informed action selection
  • Offline reinforcement learning with Decision Transformer architecture for CAD sequence reconstruction
  • Geometric alignment rewards to incorporate continuous visual feedback for a human-like modeling process

Methodology

SOV-CAD formulates CAD reconstruction as a sequential decision-making task, using offline RL with a Decision Transformer. At each step, it receives stepwise orthographic views (target, incrementally built model, active sketch) and uses geometric alignment rewards to guide the model.

Key Results

SOV-CAD surpasses state-of-the-art methods in CAD sequence reconstruction and exhibits strong data efficiency.

Tags