直接从点云预测3D建筑线框，无需先找顶点。
S23DR 2026: End-to-End 3D Wireframe Prediction via DETR-Style Set Prediction with Contrastive Denoising
- 用类似DETR的集合预测方式，直接输出边坐标对。
- 对比去噪训练让早期匹配更稳定，提升精度至HSS 0.575。
- 适合做3D结构重建、自动驾驶场景理解的研究者。
我们提出WireframeDETR，作为结构化语义3D重建（S23DR）2026挑战赛的参赛方法，旨在从多视角COLMAP点云中预测3D建筑线框。该方法将DETR式的集合预测直接应用于3D点云，以边坐标对的形式输出线框，跳过中间顶点检测阶段。提出三项技术贡献：(1) 对比去噪训练，稳定早期训练中的匈牙利匹配；(2) 多尺度编码器，通过可学习标量权重融合最后一层编码器输出；(3) 逐步辅助损失加权，使梯度信号集中于最受益的解码器层。模型在公开测试集上取得HSS 0.575（F1≈0.664，IoU≈0.516），在清理后的验证集上最佳验证成绩为HSS 0.534。
原文摘要 · Abstract (English)
We present WireframeDETR, our submission to the Structured Semantic 3D Reconstruction (S23DR) 2026 Challenge, which requires predicting a 3D building wireframe from multi-view COLMAP point clouds. Our method applies DETR-style set prediction directly to 3D point clouds, producing wireframes as sets of edge coordinate pairs without any intermediate vertex detection stage. We introduce three technical contributions: (1) contrastive denoising training that stabilises noisy Hungarian matching in early epochs; (2) a multi-scale encoder that aggregates the last encoder layer outputs via learned scalar weights; and (3) progressive auxiliary loss weighting that concentrates gradient signal on the decoder layers that most benefit from it. Our model achieves a public test HSS of 0.575 (F1~=~0.664, IoU~=~0.516) and a best validation HSS of 0.534 on the cleaned val split.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。