MS-RAFT-3D: A Multi-Scale Architecture for Recurrent Image-Based Scene Flow

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 atthis https URL.

@article{schmid2025_2506.01443,
  title={ MS-RAFT-3D: A Multi-Scale Architecture for Recurrent Image-Based Scene Flow },
  author={ Jakob Schmid and Azin Jahedi and Noah Berenguel Senn and Andrés Bruhn },
  journal={arXiv preprint arXiv:2506.01443},
  year={ 2025 }
}