Vasha DuTell

Dr. Vasha DuTell is a Postdoctoral Research Associate at the Laboratory for Atmospheric Experiments (LAE) and Lecturing Faculty at UC Berkeley's School of Information. She holds a PhD in Vision Science from UC Berkeley, where she collaborated with Drs. Bruno Olshausen and Martin Banks on natural scene statistics and human vision modeling. Her postdoctoral work at MIT CSAIL focused on spatiotemporal and peripheral vision modeling under Drs. Ruth Rosenholtz and Bill Freeman.

Her research combines computer vision and deep learning to address environmental challenges like contrail detection from satellite and ground-based imagery. She develops segmentation models for contrail tracking and studies their environmental impacts. Vasha also teaches computer vision at UC Berkeley, bridging academic instruction with cutting-edge research.

Her recent publications span topics such as peripheral vision modeling, perceptual straightness in machine learning models, and high-fidelity tracking systems. These studies emphasize interdisciplinary approaches, merging computational methods with biological insights to advance vision science and environmental monitoring.

Dr. DuTell’s work has contributed to foundational datasets like Coco-periph and frameworks like GramStatTexNet, which explore human-machine perception gaps and texture model statistics. Her efforts aim to enhance both theoretical understanding and practical applications in vision systems and climate research.