提出QAL损失函数，提升3D重建的召回与精度平衡
QAL: A Loss for Recall Precision Balance in 3D Reconstruction
- 用覆盖加权与未覆盖真值吸引项解耦召回与精度
- 在多个任务中平均提升4.3分（相比CD），恢复细长结构
- 适合需要高可靠性3D重建的机器人抓取等场景
体素学习支撑诸多3D视觉任务，如补全、重建和网格生成，但训练目标仍依赖于切比雪夫距离（CD）或地球移动距离（EMD），难以平衡召回与精度。本文提出质量感知损失（QAL），可直接替代CD/EMD，结合覆盖加权最近邻项与未覆盖真值吸引项，将召回与精度显式解耦为可调组件。在多种管道中，QAL实现一致的覆盖增益，平均比CD提升4.3分，比最优替代方案提升2.8分。尽管百分比提升较小，但显著恢复了细长结构与低代表区域。大量消融实验表明其在超参数与输出分辨率上均表现稳定；在PCN与ShapeNet上的完整重训练验证了跨数据集与骨干网络的泛化能力。此外，使用QAL训练的补全结果在GraspNet评估中获得更高抓取得分，表明覆盖性提升可直接转化为更可靠的机器人操作性能。QAL提供了一种原理清晰、可解释且实用的目标函数，适用于鲁棒3D视觉与安全关键型机器人流程。
原文摘要 · Abstract (English)
Volumetric learning underpins many 3D vision tasks such as completion, reconstruction, and mesh generation, yet training objectives still rely on Chamfer Distance (CD) or Earth Mover's Distance (EMD), which fail to balance recall and precision. We propose Quality-Aware Loss (QAL), a drop-in replacement for CD/EMD that combines a coverage-weighted nearest-neighbor term with an uncovered-ground-truth attraction term, explicitly decoupling recall and precision into tunable components. Across diverse pipelines, QAL achieves consistent coverage gains, improving by an average of +4.3 pts over CD and +2.8 pts over the best alternatives. Though modest in percentage, these improvements reliably recover thin structures and under-represented regions that CD/EMD overlook. Extensive ablations confirm stable performance across hyperparameters and across output resolutions, while full retraining on PCN and ShapeNet demonstrates generalization across datasets and backbones. Moreover, QAL-trained completions yield higher grasp scores under GraspNet evaluation, showing that improved coverage translates directly into more reliable robotic manipulation. QAL thus offers a principled, interpretable, and practical objective for robust 3D vision and safety-critical robotics pipelines
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。