S23DR 2026 Winning Solution
Not on the map. Why?
Abstract (arXiv)
This text presents the winning solution to the S23DR 2026 challenge for structured 3D wireframe reconstruction from sparse SfM, fitted depth, and semantic segmentations. The method treats vertices as a conditional set and denoises 64 vertex tokens with a flow-matching DiT conditioned on Perceiver-style scene tokens. A global pass predicts the coarse structure, a hull-cropped second pass refines it, and a small multi-sample consensus step keeps the stochastic sampler well behaved. The final system ranked first on the private leaderboard, achievingHSS = 0.654.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 4 of its 7 references that are in the map. First posted on arXiv 4 Jun 2026.
Builds on
4 of 7 references.
References · 4
- 2022Scalable Diffusion Models with Transformerscited by 6,054
- 2022Flow Matching for Generative Modelingcited by 5,702
- 2021Perceiver: General Perception with Iterative Attentioncited by 1,104
- 2025Explaining Human Preferences via Metrics for Structured 3D Reconstructioncited by 3
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2026Reconstruction of a 3D wireframe from a single line drawing via generative depth estimation
- 2024Sp2360: Sparse-view 360 Scene Reconstruction using Cascaded 2D Diffusion Priors
- 2026S-VGGT: Structure-Aware Subscene Decomposition for Scalable 3D Foundation Models
- 2024Pragmatist: Multiview Conditional Diffusion Models for High-Fidelity 3D Reconstruction from Unposed Sparse Views
- 2026Speed3R: Sparse Feed-forward 3D Reconstruction Models
- 2025CLR-Wire: Towards Continuous Latent Representations for 3D Curve Wireframe Generation
- 2024MVSplat360: Feed-Forward 360 Scene Synthesis from Sparse Views
- 2026Towards Valid B-Rep Generation: Training-Free Wireframe Anomaly Detection and Repair