معرفی
Bardia Doosti is a Researcher at Google AR/VR. He holds a PhD in Computer Science from Indiana University Bloomington (2021), where he worked under Prof. David Crandall at the IU Computer Vision Lab. His research focuses on applying Deep Learning to Computer Vision and Audio Signal Processing, with emphasis on 3D Reconstruction, Hand and Object Pose Estimation, and Graph Neural Networks.
His publications span topics like Egocentric Vision, Material Design Analysis, and Sound Localization. Key contributions include the HOPE-Net and Spectral Graphormer models. He has collaborated with researchers at Google AI, Facebook Reality Labs, and Indiana University. His work leverages large datasets like Rico for UI pattern analysis and web design evolution studies using Hidden Markov Models and GANs.
