R. Deebalakshmi

Papers

Total Citations

H-Index

About

R. Deebalakshmi is a researcher at the forefront of computer vision and video processing, with a particular focus on optimizing deep learning models for modern video codecs. Her most-cited work introduces a novel, optimized YOLO model specifically designed for H.265 encoded video frames, addressing the critical challenge of object detection in compressed video streams. This contribution is significant because H.265 (HEVC) is widely used for high-efficiency video compression, yet its encoding artifacts often degrade detection accuracy. By tailoring YOLO—a state-of-the-art real-time object detection framework—to this domain, Deebalakshmi’s research enables more robust and efficient analysis of surveillance, streaming, and autonomous vehicle footage. Her work has already garnered attention, with 4 citations in its first year, signaling growing impact in the field. Deebalakshmi’s research bridges the gap between video compression standards and AI-driven analytics, offering practical solutions for real-world applications where bandwidth and accuracy are both critical. Her innovative approach marks her as a promising voice in the evolving landscape of intelligent video systems.