Sena Kiciroglu

École Polytechnique Fédérale de Lausanne

Papers

Total Citations

H-Index

About

Sena Kiciroglu is a researcher advancing the field of human motion prediction, a critical component for safety in autonomous driving and human-robot interaction. Her work centers on developing deep learning models that anticipate future human poses, with a particular focus on long-term forecasting. Kiciroglu’s most influential contribution, "Motion Prediction Using Temporal Inception Module" (2021), has garnered 41 citations, introducing a novel architecture that effectively exploits different temporal scales to improve prediction accuracy across varying input lengths. She further innovated in her 2022 work on "Long Term Motion Prediction Using Keyposes," demonstrating that predicting every instantaneous pose is unnecessary; instead, forecasting only key poses yields more effective long-term results. This insight challenges conventional sequence-to-sequence approaches by prioritizing efficiency and relevance. With a total of 55 citations across her top papers, Kiciroglu’s research is shaping how machines understand and anticipate human movement, offering practical solutions for real-time, safety-critical applications. Her work bridges the gap between theoretical deep learning and tangible robotic systems, making her a notable voice in the intersection of computer vision and robotics.