Agnieszka Siwocha
Papers
1
Total Citations
20
H-Index
1
About
Agnieszka Siwocha is a researcher at the forefront of computer vision, with a focused expertise in human pose estimation and deep learning architectures. Her most impactful work, "Combined YOLOv5 and HRNet for High Accuracy 2D Keypoint and Human Pose Estimation" (2022), has garnered 20 citations, demonstrating its influence in the field. In this study, Siwocha innovatively integrates YOLOv5’s efficient object detection with HRNet’s high-resolution representation learning, achieving superior accuracy in 2D keypoint localization. This hybrid approach addresses critical challenges in real-world applications, from sports analytics and medical fall detection to human-robot interaction, where precise pose estimation is essential. By building upon foundational work by Li et al. at CVPR 2020, Siwocha advances the state of the art, offering a practical solution that balances speed and precision. Her contributions are particularly notable for bridging detection and pose estimation pipelines, making her research highly relevant for students and practitioners seeking robust, deployable models. With a clear trajectory in enhancing CNN-based systems, Siwocha is a rising voice in applied computer vision, driving progress toward more accurate and responsive human-aware technologies.
Research Focus
Key Achievements
Top Papers
- 1