Iris Walter
Papers
2
Total Citations
19
H-Index
2
About
Iris Walter is a researcher at the forefront of efficient deep learning for embedded and assistive robotic systems. Her work centers on making convolutional neural networks (CNNs) practical for resource-constrained devices, with a particular focus on real-time face recognition and personalized human-robot interaction. Walter’s most impactful contribution is the development of CNNParted, an open-source framework that enables efficient CNN inference partitioning across heterogeneous embedded platforms. This work, which has garnered 11 citations since 2023, addresses a critical bottleneck in deploying AI on edge devices by optimizing computational load and latency. Her earlier research on embedded face recognition for assistive robotics, cited 8 times, demonstrates how personalized services can be delivered through low-power vision systems, directly enhancing the autonomy and responsiveness of care robots. By bridging the gap between high-performance deep learning and the strict constraints of embedded hardware, Walter is enabling smarter, more accessible robotic assistants. Her open-source contributions further amplify her impact, providing a foundation for other researchers to build upon in the growing field of edge AI.
Research Focus
Key Achievements
Top Papers
- 1
- 2