将多尺度设计引入图像场景流，显著提升精度。
MS-RAFT-3D: A Multi-Scale Architecture for Recurrent Image-Based Scene Flow
- 构建分层特征与上下文编码的多尺度递归框架
- 在KITTI和Spring数据集上分别提升8.7%和65.8%
- 适合做3D运动估计与自动驾驶相关研究者
尽管多尺度设计在光流和立体匹配的递归网络中表现优异，但在基于图像的场景流领域尚未被采用。为此，我们基于单尺度递归场景流主干网络，提出一种多尺度方法，将光流中的层次化思想成功迁移至图像场景流。通过合理设计特征编码器、上下文编码器、整体粗到细框架及训练损失函数，所提方法在KITTI数据集上达到3.89的误差（优于当前最优4.26），在Spring数据集上达到9.13的误差（优于当前最优26.71），分别提升8.7%和65.8%。代码已开源：https://github.com/cv-stuttgart/MS-RAFT-3D。
原文摘要 · Abstract (English)
Although multi-scale concepts have recently proven useful for recurrent network architectures in the field of optical flow and stereo, they have not been considered for image-based scene flow so far. Hence, based on a single-scale recurrent scene flow backbone, we develop a multi-scale approach that generalizes successful hierarchical ideas from optical flow to image-based scene flow. By considering suitable concepts for the feature and the context encoder, the overall coarse-to-fine framework and the training loss, we succeed to design a scene flow approach that outperforms the current state of the art on KITTI and Spring by 8.7%(3.89 vs. 4.26) and 65.8% (9.13 vs. 26.71), respectively. Our code is available at https://github.com/cv-stuttgart/MS-RAFT-3D.
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。