Daniel Bezerra

About

Daniel Bezerra is a researcher at the forefront of efficient deep learning for constrained environments, with a primary focus on computer vision and edge AI. His most impactful work, "FCN-Pose: A Pruned and Quantized CNN for Robot Pose Estimation for Constrained Devices" (2022, 20 citations), addresses a critical challenge in the Internet of Things (IoT) ecosystem: deploying computationally intensive deep learning models on resource-limited devices. Bezerra's major contribution lies in demonstrating that advanced model compression techniques—specifically pruning and quantization—can enable accurate robot pose estimation on hardware with severe constraints on processor, RAM, and storage. This work bridges the gap between state-of-the-art computer vision algorithms and practical IoT deployment, making real-time robotics more accessible. By tackling the inherent tension between model performance and device limitations, Bezerra has provided a blueprint for deploying deep learning on edge devices, a key enabler for smart manufacturing, autonomous systems, and ubiquitous computing. His research continues to shape how we think about AI efficiency, ensuring that even the smallest devices can leverage powerful neural networks.

Research Focus

Key Achievements

Top Papers

1. 1
   FCN-Pose: A Pruned and Quantized CNN for Robot Pose Estimation for Constrained Devices

   20 citations · 2022

Key Collaborators

Contact & Links