[论文解读] Density-aware Chamfer Distance as a Comprehensive Metric for Point Cloud Completion
引入 Density-aware Chamfer Distance (DCD),一种界限化、密度敏感的度量方法,源自 Chamfer Distance,用于更好地评估和训练点云完成,并辅以点判别器和引导下采样。
Chamfer Distance (CD) and Earth Mover's Distance (EMD) are two broadly adopted metrics for measuring the similarity between two point sets. However, CD is usually insensitive to mismatched local density, and EMD is usually dominated by global distribution while overlooks the fidelity of detailed structures. Besides, their unbounded value range induces a heavy influence from the outliers. These defects prevent them from providing a consistent evaluation. To tackle these problems, we propose a new similarity measure named Density-aware Chamfer Distance (DCD). It is derived from CD and benefits from several desirable properties: 1) it can detect disparity of density distributions and is thus a more intensive measure of similarity compared to CD; 2) it is stricter with detailed structures and significantly more computationally efficient than EMD; 3) the bounded value range encourages a more stable and reasonable evaluation over the whole test set. We adopt DCD to evaluate the point cloud completion task, where experimental results show that DCD pays attention to both the overall structure and local geometric details and provides a more reliable evaluation even when CD and EMD contradict each other. We can also use DCD as the training loss, which outperforms the same model trained with CD loss on all three metrics. In addition, we propose a novel point discriminator module that estimates the priority for another guided down-sampling step, and it achieves noticeable improvements under DCD together with competitive results for both CD and EMD. We hope our work could pave the way for a more comprehensive and practical point cloud similarity evaluation. Our code will be available at: https://github.com/wutong16/Density_aware_Chamfer_Distance .
研究动机与目标
- 识别 Chamfer Distance (CD) 和 Earth Mover’s Distance (EMD) 在点云完成评估中的局限性。
- 提出 Density-aware Chamfer Distance (DCD),在保持高效和有界的同时捕捉密度分布与局部细节。
- 证明 DCD 作为可靠的评估指标和有效的训练损失函数。
- 开发一个点判别器模块,以实现完成阶段的引导下采样和点分布的平衡。
提出的方法
- 通过对 CD 进行一阶泰勒展开来推导 DCD,以将每点的贡献界定在 0 与 1 之间。
- 通过用其查询频次 n_y 和一个温度参数 alpha 对每一点的贡献进行加权,纳入密度感知项。
- 引入一个由 DCD 构成的双项对称损失,以提供有界、梯度稳定的优化。
- 设计一个神经点判别器,预测一个密度感知信号 g(x) 以识别点的重要性。
- 利用判别器输出进行引导下采样,在两阶段完成中减少离群点并平衡密度。
- 在训练中可选地用 DCD 损失替代 CD 损失,以改善全局结构和局部细节。
实验结果
研究问题
- RQ1密度感知型 Chamfer Distance 的修改是否能在不同局部密度和噪声水平下提供更一致的评估?
- RQ2DCD 作为评估指标和训练损失是否在点云完成任务中优于 CD 与 EMD?
- RQ3密度感知判别器加上引导下采样是否能在跨数据集上提高完成质量和稳定性?
- RQ4在常见的 PCN 家族架构上使用 DCD 作为损失函数的实际效果是什么?
- RQ5与 CD 和 EMD 相比,DCD 如何在保持全局结构与细节之间取得平衡?
主要发现
- DCD 提供有界、[0,1]-归一化的距离,与全局结构和局部几何都相关。
- DCD 比 CD 和 EMD 更一致、可靠,尤其在 CD 与 EMD 存在分歧时。
- 以 DCD 作为损失进行训练可以减少 EMD,也能相较于使用 CD 训练的模型减少 CD。
- 一个包含密度感知点判别器和引导下采样的两阶段完成框架可获得更好的视觉质量和点分布平衡。
- 在竞争基线中用 DCD 换代 CD 损失,在 CD、EMD、DCD 指标之间实现有利的权衡。
更好的研究,从现在开始
从阅读论文到最终审阅,大幅缩短您的研究时间。
无需绑定信用卡
本解读由 AI 生成,并经人工编辑审核。