← 研踪 PaperTrailarXiv:2609.12557 · v2
DRS-VPT: Directly Relocalizing in a Scan with Vision Point Transformers
首次提交 2026-09-11 16:02
版本更新 2026-11-11 05:41
雷达观测 2026-11-13 10:09
摘要
We present DRS-VPT, a feed-forward transformer architecture for foundational image-to-scan registration. Given query images and a reference 3D point cloud, the model predicts the scan pose and point map alongside the poses and point maps of each camera, all expressed in the first camera's frame. It additionally predicts a coarse-to- fine pyramid of per-point and per-pixel features for direct reprojective alignment of the scan to the first image. This formulation unifies downstream tasks such as camera-LiDAR calibration in autonomous driving and indoor camera-to-map relocalization. A single DRS-VPT model achieves state-of-the-art performance for image-to-LiDAR registration in autonomous driving, competitive indoor relocalization without training map-specific weights, and strong zero-shot transfer to unseen environments. We also show qualitatively that the model learns complex scan-to-image projection properties such as occlusion of back-facing points. Code and models are available at https://github.com/ori-drs/drs-vpt.
作者备注：ECCV 2026. Code: https://github.com/ori-drs/drs-vpt Project page: https://ori-drs.github.io/drs-vpt-pages/
观测到的版本
相邻记录可能跳过中间版本；首次发现 v2 及以上也会标为版本更新。
v2 · 2026-11-11 05:41 · 最新观测版本
DRS-VPT: Directly Relocalizing in a Scan with Vision Point Transformers
We present DRS-VPT, a feed-forward transformer architecture for foundational image-to-scan registration. Given query images and a reference 3D point cloud, the model predicts the scan pose and point map alongside the poses and point maps of each camera, all expressed in the first camera's frame. It additionally predicts a coarse-to- fine pyramid of per-point and per-pixel features for direct reprojective alignment of the scan to the first image. This formulation unifies downstream tasks such as camera-LiDAR calibration in autonomous driving and indoor camera-to-map relocalization. A single DRS-VPT model achieves state-of-the-art performance for image-to-LiDAR registration in autonomous driving, competitive indoor relocalization without training map-specific weights, and strong zero-shot transfer to unseen environments. We also show qualitatively that the model learns complex scan-to-image projection properties such as occlusion of back-facing points. Code and models are available at https://github.com/ori-drs/drs-vpt.
查看此版本解读 →查看此版本原文 ↗