Papers
3
Total Citations
11
H-Index
2
About
Juan Tarrio is a robotics researcher specializing in visual simultaneous localization and mapping (SLAM), with a focus on integrating semantic understanding and structural edge information into real-time navigation systems. His most impactful work, "SeMLaPS: Real-Time Semantic Mapping With Latent Prior Networks and Quasi-Planar Segmentation" (2023, 6 citations), introduces a novel methodology that combines 2D and 3D neural networks to enable real-time semantic mapping from RGB-D sequences, significantly enhancing the geometric functionality of SLAM systems for robotic and AR/VR applications. Tarrio also contributed to loop closure detection with his 2018 paper on edge-based methods, addressing error drift in visual SLAM, and developed SE-SLAM (2019), a semi-dense structured edge-based monocular SLAM system that bridges the gap between sparse feature-based methods and direct approaches. His work demonstrates a consistent focus on improving map density and semantic richness while maintaining real-time performance, making his contributions particularly relevant for autonomous navigation and augmented reality. Though early in his career, Tarrio’s research is already shaping how robots perceive and interact with their environments.
Research Focus
Key Achievements
Top Papers
- 1
- 2Edge-Based Loop Closure Detection in Visual SLAM3 citations · 2018
- 3SE-SLAM: Semi-Dense Structured Edge-Based Monocular SLAM2 citations · 2019