用合成数据与两阶段模型，提升作物3D重建与性状提取精度
Denoising-Aware Temporal Point Cloud Completion for 3D Crop Architecture Recovery and Phenotypic Trait Extraction
- 先去噪再时序补全，利用前后生长阶段信息恢复遮挡区域
- 在合成数据上达0.0061的Chamfer距离，在真实数据上F-Score达0.2080
- 适合高通量作物表型分析研究者，支持性状自动提取
高通量表型依赖于植物在不同生长期的精确3D重建，但受限于缺乏具有完整几何真值的数据集，时序补全方法的发展与评估进展缓慢。为此，我们提出SynthCrop4D——一个程序生成的、随时间演化的植物点云合成数据集，可控制噪声、遮挡及完整植物几何结构，用于重建方法的基准测试。基于该数据集，我们评估了一种两阶段流程：首先通过去噪模块消除激光扫描点云中的结构伪影，随后使用自适应时序PoinTr模型，结合前一时相(t-1)信息重建当前时相(t)的完整形态，实现因自遮挡缺失区域的恢复。我们在SynthCrop4D和真实世界Pheno4D数据集（番茄与玉米）上评估了该框架，包含去噪与不去噪两种设置。结果表明，去噪显著提升重建质量：最佳配置在SynthCrop4D上达到0.0061的Chamfer Distance，Pheno4D上F-Score为0.2080。我们进一步验证了补全后点云在性状提取中的应用，包括株高、冠层宽度与凸包体积，合成数据下凸包体积平均绝对误差为0.021，真实数据为0.343。综上，SynthCrop4D与所提流程为时序植物重建与高通量作物表型分析提供了基准与方法支持。
原文摘要 · Abstract (English)
High-throughput phenotyping depends on accurate 3D reconstruction of plants across growth stages, yet the development and evaluation of temporal completion methods are limited by the lack of datasets with complete geometric ground truth. To address this challenge, we introduce SynthCrop4D, a procedurally generated synthetic dataset of temporally evolving plant point clouds that provides controllable noise, occlusion, and complete plant geometry for benchmarking reconstruction methods. Using this dataset, we evaluate a two-stage pipeline that combines spatial denoising and temporal point cloud completion. First, a denoising module removes structural artifacts from raw laser-scanned point clouds. The resulting data are then processed by an Adaptive Temporal PoinTr model that reconstructs the current growth stage (t) using information from the previous stage (t-1), enabling recovery of regions missing due to self-occlusion. We evaluate the proposed framework on both SynthCrop4D and the real-world Pheno4D dataset (tomato and maize) under settings with and without denoising. Results show that denoising substantially improves reconstruction quality, with the best configuration achieving a Chamfer Distance of 0.0061 on SynthCrop4D (Temporal PoinTr + Mamba-DG) and an F-Score of 0.2080 on Pheno4D (Vanilla PoinTr + Mamba-DG). We further demonstrate the use of completed point clouds for phenotypic trait extraction, including plant height, canopy width, and convex hull volume, obtaining hull-volume MAEs of 0.021 on synthetic data and 0.343 on real data. Together, SynthCrop4D and the proposed pipeline provide a benchmark and methodology for temporal plant reconstruction and high-throughput crop phenotyping.
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。