Junying Zeng

Papers

Total Citations

H-Index

About

Junying Zeng is a researcher whose work sits at the intersection of computer vision and affective computing, with a particular focus on advancing facial expression recognition (FER) under challenging, real-world conditions. Their most notable contribution, the 2019 paper "Weakly supervised facial expression recognition via transferred DAL-CNN and active incremental learning," has garnered 18 citations, establishing a foundation for more efficient and adaptable emotion recognition systems. This work introduces a novel approach that leverages transfer learning and a discriminative active learning framework (DAL-CNN) to overcome the limitations of heavily labeled datasets, enabling models to learn effectively from limited or noisy supervision. By incorporating active incremental learning, Zeng’s method allows systems to continuously improve as new data becomes available, a crucial step toward deploying FER in dynamic environments like human-robot interaction or mental health monitoring. Their research bridges the gap between supervised and unsupervised learning, offering a practical solution for scalable emotion analysis. For students and researchers in machine learning and human-centered AI, Junying Zeng’s work represents a thoughtful blend of algorithmic innovation and real-world applicability, pushing the boundaries of how machines understand human expressions.