GRF-Recon: Global Ray-Field Optimization for Long-Sequence Feed-forward Reconstruction

2026-09-17

AI Summary

该研究针对长序列单目图像3D重建中的内存占用、几何精度和轨迹漂移问题，提出了一种结合粗到细轨迹对齐与轻量级几何先验注入的统一框架，并通过混合权重稀疏光线场优化来增强跨帧几何一致性。

Abstract

Feed-forward 3D reconstruction provides an efficient paradigm for scene modeling from image sequences. Scaling these models to large monocular scenarios are constrained by excessive GPU memory footprint, degraded local geometry, and long-term trajectory drift. Existing chunk-based optimization strategies provide limited geometric constraints and fail to maintain global consistency over extended trajectories. We present a unified framework for stable and scalable feed-forward 3D reconstruction from long monocular sequences. Our approach builds on coarse-to-fine trajectory alignment augmented by lightweight geometric prior injection. Distilling monocular geometric cues into the feed-forward backbone via LoRA adaptation improves depth accuracy on fine structures while preserving inference efficiency. We introduce a hybrid-weight sparse ray-field optimization that leverages high-frequency geometric features to guide local point-cloud refinement and enforce consistent inter-frame ray constraints. Unlike prior chunk-based methods, this establishes strong cross-frame geometric coupling while maintaining scalability. Finally, an efficient trajectory stitching strategy with joint ray-error optimization explicitly reduces accumulated drift. Extensive experiments show that our approach achieves competitive trajectory accuracy compared with representative SLAM systems, while maintaining globally consistent 3D reconstruction in large-scale scenarios.

Problem

• feed-forward 3D reconstruction
• long monocular sequences
• GPU memory footprint
• local geometry degradation
• trajectory drift

Innovation

• coarse-to-fine trajectory alignment
• lightweight geometric prior injection
• hybrid-weight sparse ray-field optimization
• trajectory stitching strategy