work MultiModal (2D/3D) Diffusion Models Work on 2D/3D diffusion model framework that offers high fidelity, controllability, modularity, (re)-usability (adapting existing foundational models) and applicability (data augmentation). Scalable Channel Mixer for Vision Transformers Generic channel mixer to scale all ViTs Action Recognition and Localization Advance the SOTA for action recognition and detection in videos Parameter and State Estimation in Linear Time Varying Systems Master's Thesis
