Charika De Alvis

About

Charika De Alvis is a researcher whose work lies at the intersection of robotics, computer vision, and scene understanding, with a particular focus on leveraging multi-modal sensor data for autonomous navigation. Her key research areas include urban scene segmentation, conditional random fields (CRFs), and online learning for robotic perception. De Alvis made significant contributions by developing methods that integrate data from diverse sensors—such as colour cameras and 3D laser scanners—to improve scene labelling accuracy. Her 2016 paper, "Urban scene segmentation with laser-constrained CRFs," introduced a novel framework that uses laser range data to constrain CRF models, achieving more robust segmentation in complex urban environments. This work, cited 5 times, laid the groundwork for her subsequent 2017 study on online learning for scene segmentation, which addressed the challenge of adapting models to unknown environments in real time. Though her citation counts are modest, De Alvis’s research is notable for its practical focus on enabling robots to navigate unfamiliar spaces with greater autonomy and reliability, making her a promising voice in the field of embodied AI.

Research Focus

Key Achievements

Top Papers

1. 1
   Urban scene segmentation with laser-constrained CRFs

   5 citations · 2016
2. 2
   Online learning for scene segmentation with laser-constrained CRFs

   2 citations · 2017

Key Collaborators

Contact & Links