• 2026

This work proposes a novel pipeline for real-time monocular V-SLAM, SLAM-MER, that (1) introduces a new spatio-temporal camera localization approach, and (2) achieves real-time performance (>80 FPS). The framework, developed from scratch in C++, is open-source and modular to allow further V-SLAM development.

• 2025

SAC-GNC is a novel adaptive annealing strategy for Graduated Non-Convexity (GNC) that integrates sample consensus into the classical GNC framework. The addition of sample consensus improves robustness and reduces dependence on predefined parameters. The effectiveness of this approach is demonstrated in point cloud registration and pose graph optimization tasks.

• 2024

We propose a novel pixel sampler for neural implicit rendering that leverages the scene's Signed Distance Function (SDF). This approach achieves sharper reconstructions and improved performance in less-observed regions without requiring additional prior data.

• 2023

BANSAC is a novel sampling strategy for RANSAC. We introduce a dynamic Bayesian network that iteratively updates the inlier probabilities of individual data points. In each iteration, these updated probabilities guide the sampling process. This approach outperforms state-of-the-art baselines in terms of both accuracy and efficiency.