RIDE: Relocalization-Informed Depth Estimation with 3D Gaussian Splatting

2026-09-10

AI Summary

RIDE利用3D高斯点云模型和PnP-RANSAC对应关系，结合预训练视频深度模型，从RGB流中估计密集度量深度，提高深度准确性和时间一致性。

Abstract

Render--match--PnP relocalization establishes correspondences between query image pixels and 3D map points for camera pose recovery, but their potential to support dense depth estimation is often overlooked. To exploit this geometric information, we present RIDE, which estimates dense metric depth from a robot's RGB stream. Given a metrically scaled 3D Gaussian Splatting (3DGS) model, RIDE combines sparse metric depth observations derived from PnP-RANSAC inlier correspondences with the geometric prior of a pretrained video-depth model. To handle uneven and intermittent observations, it integrates global and local depth correction with temporal memory, supporting depth estimation through short observation gaps after metric scale initialization. Trained on public RGB-D videos, RIDE is evaluated on robot sequences without fine tuning. Experiments show improved depth accuracy and temporal consistency over scale-only calibration, demonstrating how localization geometry can support both pose recovery and dense robot perception.

Problem

• relocalization
• dense depth estimation
• 3D Gaussian Splatting
• PnP-RANSAC
• geometric prior

Innovation

• Relocalization-Informed Depth Estimation
• 3D Gaussian Splatting
• PnP-RANSAC
• Temporal Memory
• Dense Metric Depth