Runner-up Best Paper Award at Eurographics Computer Graphics and Visual Computing (CGVC) 2025 conference
We are pleased to announce that our paper “OCAM: Occlusion-aware Completeness via Coverage Analysis with Monte Carlo Sampling”, authored by Lizeth Joseline Fuentes Perez, Luciano Arnaldo Romero Calla, Renato Pajarola, and Javier Turek has received the Runner-up Best Paper Award at the Eurographics Computer Graphics and Visual Computing (CGVC) 2025 conference.
Incomplete reconstructions caused by occlusions are a common challenge in indoor 3D scanning. Estimating scan completeness during the scanning process can significantly improve the quality of the resulting data. OCAM introduces a lightweight, reconstruction-free framework for estimating global scan coverage. At its core lies the Blind Boundary Ray Ratio (BBR), a scalar score based on Monte Carlo ray sampling that captures whether large surface regions remain unscanned – without requiring ground-truth models, surface connectivity, or sensor trajectories.
Building on this visibility framework, OCAM also incorporates a Next View Proposal (NVP) module. By aggregating escaping rays during BBR computation, NVP suggests scanning directions that target unobserved regions, transforming OCAM from a purely passive evaluator into an actionable guidance framework for improving scan coverage.
This award highlights both the potential of OCAM to advance practical indoor 3D data acquisition and the elegance of applying computer graphics concepts to solve, in a simple manner, a challenging real-world problem.