Andrea Coifman

Papers

Total Citations

H-Index

About

Andrea Coifman’s research sits at the intersection of computer vision, human-computer interaction, and affective computing, with a primary focus on developing systems that can understand and respond to human attention and cognitive states. Her most cited work, “Subjective Annotations for Vision-based Attention Level Estimation” (2019), tackles the fundamental challenge of grounding machine learning models in reliable human judgments. By introducing a framework for collecting and utilizing subjective annotations, Coifman addresses a critical bottleneck in training robust attention estimation systems—a technology with transformative potential for human-robot interaction, driver safety monitoring, and adaptive smart home environments. While her citation count is still growing, the conceptual contribution of this work is significant: it provides a methodological foundation for creating more ecologically valid training data, moving beyond objective gaze metrics to capture the nuanced, subjective experience of attention. Coifman’s research is particularly valuable for students and practitioners working on embodied AI, as it highlights the importance of human-centered design in building systems that can truly collaborate with people. Her work exemplifies the careful, foundational thinking needed to advance the field of vision-based cognitive state estimation.