Jorge Scandaliaris
Papers
1
Total Citations
18
H-Index
1
About
Jorge Scandaliaris has made significant contributions to computer vision, particularly in robust image detection under challenging lighting conditions. His key research areas include color-based invariant feature extraction, object detection, and scene understanding. Scandaliaris is best known for his work on combining color-based invariant gradient detectors with Histogram of Oriented Gradients (HoG) descriptors, a method that dramatically improves detection accuracy in outdoor scenes plagued by cast shadows. His most-cited paper (2009, 18 citations) demonstrates how this hybrid approach outperforms traditional grayscale-based methods in urban scene classification and person detection, addressing a critical real-world problem where shadows often degrade performance. This work showcases his talent for fusing color theory with geometric features to create more resilient computer vision systems. Beyond this flagship study, Scandaliaris has explored invariant representations that maintain discriminative power across varying illumination, a challenge central to autonomous navigation and surveillance. His research has practical implications for robotics, security systems, and autonomous vehicles operating in uncontrolled environments. Through his methodical approach to solving illumination-related distortions, Scandaliaris has advanced the field's understanding of how color can be leveraged to achieve more reliable visual recognition in natural settings.
Research Focus
Key Achievements
Top Papers
- 1