# Fine-To-Coarse Global Registration of RGB-D Scans

arXiv: https://arxiv.org/abs/1607.08539 · HTML: https://mlsanity.com/paper/1607.08539/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 28 Jul 2016.

## Built on by (most cited first, 7 of 7)

- [CoFiNet: Reliable Coarse-to-fine Correspondences for Robust Point Cloud Registration](https://mlsanity.com/paper/2110.14076/) (2021)
- [REGTR: End-to-end Point Cloud Correspondences with Transformers](https://mlsanity.com/paper/2203.14517/) (2022)
- [Deep Hough Voting for Robust Global Registration](https://mlsanity.com/paper/2109.04310/) (2021)
- [Mirror3D: Depth Refinement for Mirror Surfaces](https://mlsanity.com/paper/2106.06629/) (2021)
- [3D Point Cloud Registration with Multi-Scale Architecture and Unsupervised Transfer Learning](https://mlsanity.com/paper/2103.14533/) (2021)
- [A Consistency-Aware Spot-Guided Transformer for Versatile and Hierarchical Point Cloud Registration](https://mlsanity.com/paper/2410.10295/) (2024)
- [PROFusion: Robust and Accurate Dense Reconstruction via Camera Pose Regression and Optimization](https://mlsanity.com/paper/2509.24236/) (2025)

## Abstract (arXiv)

RGB-D scanning of indoor environments is important for many applications, including real estate, interior design, and virtual reality. However, it is still challenging to register RGB-D images from a hand-held camera over a long video sequence into a globally consistent 3D model. Current methods often can lose tracking or drift and thus fail to reconstruct salient structures in large environments (e.g., parallel walls in different rooms). To address this problem, we propose a "fine-to-coarse" global registration algorithm that leverages robust registrations at finer scales to seed detection and enforcement of new correspondence and structural constraints at coarser scales. To test global registration algorithms, we provide a benchmark with 10,401 manually-clicked point correspondences in 25 scenes from the SUN3D dataset. During experiments with this benchmark, we find that our fine-to-coarse algorithm registers long RGB-D sequences better than previous methods.