PulseAugur
EN
LIVE 14:52:30

Transformers reconstruct 3D roof wireframes, win S23DR Challenge

By PulseAugur Editorial · [2 sources] ·

Researchers have developed a novel Transformer-based method for reconstructing 3D roof wireframes from sparse point clouds. This approach, inspired by DETR, dynamically subsamples input data and fuses it with semantic and Gestalt features. The system achieved a Hybrid Structure Score of 0.6476 on the "HoHo 22k" dataset, securing second place in the S23DR Challenge 2026. AI

IMPACT Introduces a novel Transformer architecture for 3D reconstruction, potentially improving scene understanding in computer vision.

RANK_REASON The cluster contains an academic paper detailing a new method and its performance on a specific dataset and challenge.

Read on arXiv cs.CV →

AI-generated summary · Google Gemini · from 2 sources. How we write summaries →

COVERAGE [2]

  1. arXiv cs.CV TIER_1 English(EN) · Gustav Hanning, Ludvig Dill\'en, Jonathan Astermark, Johanna Lidholm, Viktor Larsson ·

    Edge Prediction for Roof Wireframe Reconstruction with Transformers

    arXiv:2606.02406v1 Announce Type: new Abstract: This paper presents a competitive solution to the S23DR Challenge 2026, which aims to reconstruct 3D house roof wireframe models from sparse SfM point clouds and ground-level semantic segmentations and depth maps. Our proposed metho…

  2. arXiv cs.CV TIER_1 English(EN) · Viktor Larsson ·

    Edge Prediction for Roof Wireframe Reconstruction with Transformers

    This paper presents a competitive solution to the S23DR Challenge 2026, which aims to reconstruct 3D house roof wireframe models from sparse SfM point clouds and ground-level semantic segmentations and depth maps. Our proposed method utilizes an end-to-end Transformer encoder-dec…

RELATED ENTITIES

RELATED TOPICS