NoDrift3R: Raymap-Guided Coupling for Drift-Robust Unposed Feed-Forward 3D Reconstruction

2026-07-08

AI Summary

This work addresses the geometric inaccuracies and degraded rendering quality in pose-free feedforward 3D reconstruction from long image sequences, which stem from camera pose drift. To mitigate this issue, the authors propose a geometry-appearance co-optimization framework that introduces a novel Raymap-guided coupling module to enable bidirectional feedback between geometry and appearance. Temporal stability is further enhanced through a dual-frequency view scheduling strategy. By jointly optimizing 3D Gaussian splatting, raymap consistency, RGB reconstruction, and camera regularization, the method effectively suppresses pose drift. Experiments demonstrate that the approach significantly improves reconstruction accuracy and rendering fidelity on both in-domain and cross-domain long-sequence scenarios, exhibiting strong robustness and superior drift resistance.

Abstract

Pose-Free Feed-forward 3D Gaussian Splatting (3DGS) has recently emerged as a powerful paradigm for fast scene reconstruction. However, its performance degrades significantly in long image sequences due to cumulative camera pose estimation drift, which propagates errors into geometric modeling and severely limits rendering fidelity. In this work, we revisit the long-sequence bottleneck and identify pose drift as the primary factor restricting reconstruction quality. Furthermore, while SfM-based pseudo ground-truth poses introduce sensor noise, purely rendering-based supervision often leads to optimization instability and local minima due to the entangled optimization of geometry and pose. To address the challenges, we propose a synergistic pose-free framework that explicitly couples geometry and appearance via a Raymap-Guided Coupling Module (RGC). Concretely, we anchor Gaussian centers to raymap-induced geometry and jointly optimize RGB reconstruction, raymap consistency, and camera regularization under a unified objective, yielding a bidirectional feedback loop: stronger geometry improves rendering, and appearance supervision in turn refines geometry and pose. To further stabilize learning across wide temporal ranges, we introduce a Dual-Frequency Viewpoint Scheduling strategy that combines easy-to-hard interval expansion with replay of short-interval pairs. Extensive experiments across in-domain and cross-domain datasets show consistent gains in both rendering and pose estimation, with notably improved robustness on long sequences. Ablation studies validate our central insight: explicitly designed geometry-appearance synergy is the key to scalable and drift-robust pose-free feed-forward 3D reconstruction.

Problem

• pose drift
• 3D reconstruction
• camera pose estimation
• long image sequences
• rendering fidelity

Innovation

• Raymap-Guided Coupling
• Pose-Free 3D Reconstruction
• Drift-Robust
• 3D Gaussian Splatting
• Geometry-Appearance Synergy