Sotirios Papadopoulos

Papers

Total Citations

H-Index

About

Dr. Sotirios Papadopoulos is a computer vision researcher whose work lies at the intersection of semantic scene understanding and geometric reasoning. His most cited paper, "Semantic Image Segmentation Guided By Scene Geometry" (2021, 7 citations), introduces a novel framework that leverages depth maps to enhance the accuracy of per-pixel semantic classification in CNNs. This contribution is particularly vital for applications like autonomous driving and drone navigation, where understanding both object identity and spatial layout is critical. By demonstrating how geometric cues can refine segmentation boundaries and resolve ambiguities in cluttered scenes, Papadopoulos addresses a key limitation of purely appearance-based models. His research bridges the gap between 2D vision and 3D scene analysis, offering practical improvements for robotic perception systems. While his citation count is still growing, the targeted impact of his work on semantic segmentation—a foundational task in modern AI—positions him as an emerging voice in the field. For students and researchers, Papadopoulos’s approach exemplifies how integrating complementary data modalities can push the boundaries of what neural networks achieve in real-world environments.