Andres Marmol

Australian Centre for Robotic Vision, Queensland University of Technology, Hamburg University of Technology, Universität Hamburg

Papers

Total Citations

H-Index

About

Andrés Marmol is a computer vision and medical imaging researcher whose work sits at the intersection of surgical robotics, 3D reconstruction, and simultaneous localization and mapping (SLAM). His research has focused predominantly on developing robust visual localization and scene reconstruction systems for minimally invasive surgical environments — spaces that are notoriously difficult to navigate due to constrained anatomy, limited lighting, and uninformative imagery. Marmol's most impactful contributions center on arthroscopic surgery, where he pioneered dense 3D reconstruction techniques tailored to intra-articular environments. His flagship work, "Dense-ArthroSLAM" (2019), which has garnered 47 citations, introduced a dense reconstruction framework with robust localization priors, directly addressing the steep cognitive and physical demands placed on surgeons during joint procedures. Building on this, his earlier "ArthroSLAM" (2018) laid the multi-sensor groundwork for reliable arthroscope localization. His earlier laparoscopic research explored keyframe selection strategies and Structure from Motion approaches to improve motion estimation stability in surgical video, demonstrating a consistent commitment to making computer-assisted surgery safer and more precise. Collectively, his publications reflect a career dedicated to translating cutting-edge computer vision into clinically meaningful surgical tools.