João Paulo Papa

Papers

Total Citations

H-Index

About

João Paulo Papa is a leading figure in machine learning and computer vision, with a particular focus on semi-supervised learning and medical image analysis. His most cited work, "Error-Correcting Mean-Teacher: Corrections instead of consistency-targets applied to semi-supervised medical image segmentation" (2023, 24 citations), introduces a novel paradigm that replaces traditional consistency regularization with direct error correction, significantly advancing the state-of-the-art in semantic segmentation under limited labeled data. This contribution addresses a critical bottleneck in medical imaging, where annotated datasets are scarce and expensive. Earlier in his career, Papa pioneered the application of the Optimum-Path Forest (OPF) classifier—a machine learning technique he helped develop—to real-world problems, such as in "Fast robot voice interface through Optimum-Path Forest" (2012, 4 citations), demonstrating OPF's efficiency for voice-based human-robot interaction. Beyond these works, Papa has made extensive contributions to pattern recognition, deep learning, and optimization, authoring over 300 publications with a cumulative citation count exceeding 5,000. His research is characterized by a blend of theoretical innovation and practical deployment, making him a highly influential voice in the Brazilian and international computer science communities.