Reliability-Regulated Trajectory Optimization for Progressive COLMAP-Free 3D Gaussian Splatting

2026-09-25

AI Summary

This study addresses the trajectory drift and rendering degradation caused by early-stage error accumulation in progressive camera pose estimation for COLMAP-free 3D Gaussian Splatting. To this end, we propose a reliability-modulated self-supervised trajectory optimization framework. The core innovation lies in constructing a bidirectional cycle consistency mechanism driven by intrinsic reliability signals, which jointly governs forward motion propagation and backward trajectory correction. Coupled with sliding-window relative pose constraints, this enables progressive optimization while entirely eliminating reliance on external neural priors or offline preprocessing. Extensive experiments on the Tanks and Temples and CO3D-V2 benchmarks demonstrate that our method substantially improves both camera trajectory accuracy and novel view synthesis quality, outperforming existing pose-free baselines.

Abstract

COLMAP-free 3D Gaussian Splatting (3DGS) bypasses computationally expensive structure-from-motion (SfM) pipelines, yet progressive camera pose tracking remains fundamentally vulnerable to error compounding---early pairwise tracking inaccuracies both corrupt subsequent frame initializations and remain permanently frozen in the scene representation. Rather than relying on heavyweight external neural priors or treating progressive tracking through isolated heuristic fixes, we propose a unified reliability-regulated trajectory optimization framework for progressive COLMAP-free 3DGS. At its core, our framework establishes an intrinsic, self-supervised bidirectional cycle-consistency mechanism that systematically regulates progressive camera trajectory estimation across two complementary temporal horizons: (1) Forward Motion Propagation, where the online reliability signal adaptively gates first-order kinematic warm-starts of rigid motion into upcoming pairwise registrations, supplying informed directional search priors while safely intercepting untrusted transitions; and (2) Retrospective Trajectory Correction, where the same reliability signal dynamically weights relative-pose consistency constraints within a sliding window of neighboring camera poses. By governing both prospective state initialization and retrospective trajectory consolidation through a unified reliability regulator, our self-contained framework resolves progressive drift without external priors or offline preprocessing. Extensive evaluations on Tanks and Temples and CO3D-V2 benchmarks show that our method substantially improves camera trajectory accuracy and novel-view rendering quality, outperforming existing unposed baselines. Code is available at https://github.com/Zijian1026/RRTO-CF3DGS.

Problem

• 3D Gaussian Splatting
• COLMAP-free
• progressive camera pose tracking
• error compounding
• trajectory optimization

Innovation

• COLMAP-free 3D Gaussian Splatting
• Reliability-Regulated Trajectory Optimization
• Bidirectional Cycle-Consistency
• Progressive Camera Pose Tracking
• Self-supervised Framework