用简单网络同时实现点云升采样与去噪，速度提升47倍。
Joint Point Cloud Upsampling and Cleaning with Octree-based CNNs
- 基于改进的八叉树3D U-Net，统一处理整片点云
- 在多个基准上达到顶尖性能，推理速度提升至少47倍
- 适合追求高效简洁方案的研究者参考
从稀疏或噪声数据中恢复密集且分布均匀的点云仍是重大挑战。尽管近期进展显著，但通常依赖日益复杂的模块或繁琐的网络结构，导致推理时间长、资源消耗大。本文提倡简约设计，提出一种简单高效的联合点云升采样与去噪方法。该方法基于现成的八叉树3D U-Net（OUNet）进行小幅改进，使升采样与去噪任务在单一网络中完成。与以往按点云块处理不同，本方法直接处理整个输入点云，极大简化实现流程，并带来至少47倍的推理加速。大量实验表明，该方法在多个基准上均达到领先性能，同时具备巨大效率优势。我们期望该方法能作为简洁基线，启发研究者重新思考点云升采样与去噪的方法设计。
原文摘要 · Abstract (English)
Recovering dense and uniformly distributed point clouds from sparse or noisy data remains a significant challenge. Recently, great progress has been made on these tasks, but usually at the cost of increasingly intricate modules or complicated network architectures, leading to long inference time and huge resource consumption. Instead, we embrace simplicity and present a simple yet efficient method for jointly upsampling and cleaning point clouds. Our method leverages an off-the-shelf octree-based 3D U-Net (OUNet) with minor modifications, enabling the upsampling and cleaning tasks within a single network. Our network directly processes each input point cloud as a whole instead of processing each point cloud patch as in previous works, which significantly eases the implementation and brings at least 47 times faster inference. Extensive experiments demonstrate that our method achieves state-of-the-art performances under huge efficiency advantages on a series of benchmarks. We expect our method to serve simple baselines and inspire researchers to rethink the method design on point cloud upsampling and cleaning.
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。