Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper addresses the challenge of inaccurate depth measurements caused by glass surfaces in indoor robot navigation by proposing a training-free framework that fuses depth foundation model priors with raw sensor data. The method uses a RANSAC-based alignment to register the prior from Depth Anything with the noisy sensor depth, effectively correcting for glass-induced errors and recovering metric scale. Experiments on a newly introduced RGB-D dataset, GlassRecon, demonstrate superior performance compared to existing methods, particularly under significant depth corruption.
Glass surfaces no longer have to be a robot's kryptonite: this training-free method leverages depth priors to dramatically improve depth sensing accuracy in glass-filled environments.
Indoor robot navigation is often compromised by glass surfaces, which severely corrupt depth sensor measurements. While foundation models like Depth Anything 3 provide excellent geometric priors, they lack an absolute metric scale. We propose a training-free framework that leverages depth foundation models as a structural prior, employing a robust local RANSAC-based alignment to fuse it with raw sensor depth. This naturally avoids contamination from erroneous glass measurements and recovers an accurate metric scale. Furthermore, we introduce \ti{GlassRecon}, a novel RGB-D dataset with geometrically derived ground truth for glass regions. Extensive experiments demonstrate that our approach consistently outperforms state-of-the-art baselines, especially under severe sensor depth corruption. The dataset and related code will be released at https://github.com/jarvisyjw/GlassRecon.