| --- |
| language: |
| - en |
| tags: |
| - computer-vision |
| - 3d-reconstruction |
| - gaussian-splatting |
| - novel-view-synthesis |
| - camera-pose-estimation |
| - pose-free |
| - feed-forward |
| --- |
| |
| # NoDrift3R: Raymap-Guided Coupling for Drift-Robust Unposed Feed-Forward 3D Reconstruction |
|
|
| [**Project Page**](https://xiangyu1sun.github.io/NoDrift3R-project-page/) | [**Paper**](https://arxiv.org/abs/2607.07168) | [**Code**](https://github.com/HorizonRobotics/NoDrift3R) |
|
|
| <p align="center"> |
| <img src="https://xiangyu1sun.github.io/NoDrift3R-project-page/static/images/problem.png" alt="NoDrift3R overview" width="100%"> |
| </p> |
|
|
| NoDrift3R is an ECCV 2026 method for drift-robust, pose-free feed-forward 3D Gaussian Splatting reconstruction. It targets cumulative camera-pose drift in long, unposed image sequences by explicitly coupling geometry and appearance. Gaussian centers are anchored to raymap-induced geometry, while RGB reconstruction, raymap consistency, and camera regularization are optimized jointly to improve rendering, geometry, and pose estimation together. |
|
|
| ## Highlights |
|
|
| - **Drift-robust long-sequence reconstruction:** directly addresses cumulative camera-pose errors that degrade geometry and rendering quality. |
| - **Raymap-Guided Coupling:** connects Gaussian geometry with raymap predictions to reduce pose-geometry ambiguity. |
| - **Geometry-appearance synergy:** jointly optimizes RGB reconstruction, raymap consistency, and camera regularization in a unified objective. |
| - **Dual-Frequency Viewpoint Scheduling:** combines easy-to-hard interval expansion with replay of short-interval image pairs for stable training across temporal ranges. |
| - **Pose-free feed-forward inference:** reconstructs scenes and estimates camera poses without requiring input camera poses. |
|
|
| ## Checkpoints |
|
|
| This repository contains the following checkpoints: |
|
|
| - `dl3dv_giant_2_8_H20.ckpt` |
| - `dl3dv_large_4_8_5090.ckpt` |
| - `mix_dataset_giant_4_8_H20.ckpt` |
| - `mix_dataset_large_8_8_5090.ckpt` |
|
|
| Please follow the [GitHub repository](https://github.com/HorizonRobotics/NoDrift3R) and [project page](https://xiangyu1sun.github.io/NoDrift3R-project-page/) for code, configuration, dataset, and usage updates. |
|
|
| ## Download |
|
|
| The root `config.yaml` lists the released checkpoints with their file sizes |
| and SHA256 digests. Include it when selecting a checkpoint to download: |
|
|
| ```python |
| from huggingface_hub import snapshot_download |
| |
| checkpoint = "dl3dv_large_4_8_5090.ckpt" |
| model_dir = snapshot_download( |
| "HorizonRobotics/NoDrift3R", |
| allow_patterns=["config.yaml", checkpoint], |
| local_dir="NoDrift3R", |
| ) |
| ``` |
|
|
| Choose any checkpoint listed above and follow the |
| [NoDrift3R code](https://github.com/HorizonRobotics/NoDrift3R) for loading and |
| inference configuration. |
|
|
| ## Citation |
|
|
| If you find NoDrift3R useful in your research, please cite: |
|
|
| ```bibtex |
| @misc{sun2026nodrift3rraymapguidedcouplingdriftrobust, |
| title = {NoDrift3R: Raymap-Guided Coupling for Drift-Robust Unposed Feed-Forward 3D Reconstruction}, |
| author = {Xiangyu Sun and Liu Liu and Seungkwon Yang and Jingbing Han and Seungtae Nam and Zhizhong Su and Eunbyung Park}, |
| year = {2026}, |
| eprint = {2607.07168}, |
| archivePrefix = {arXiv}, |
| primaryClass = {cs.CV}, |
| url = {https://arxiv.org/abs/2607.07168} |
| } |
| ``` |
|
|
