Projects

Com-BrainTF - A novel local-global hierarchical transformer architecture to efficiently learn and integrate community-aware ROI embeddings for fMRI connectome analysis by utilizing both ROI-level and community-level information. Our paper got accepted in the prestigious MICCAI (The Medical Image Computing and Computer Assisted Intervention Society) 2023 conference. [Read More]

Developed DE-TensoRF, a model that can render high-quality 3D objects with as few as 3 input images, and in under 15 min on a single GPU. Proposed three techniques to achieve data-effiency: (i) Obtain unseen views using symmetry (ii) Semantic conditioning (iii) Semantic Loss. [Read More]

The Student Speaking Simulator creates an immersive room experience using projectors to replicate a live audience, and provides students with the opportunity to practice and develop their speaking skills year-round. The Simulator also records your performance and key metrics, and provides feedback for improvement. Not exclusive to public speaking, the... [Read More]

Variational seq2seq DMGNN-based GAN (V-DMGNN-GAN) model for human skeletal motion prediction. [Read More]

Built a robocode tank (a Reinforcement Learning agent) that battles with another robot (an enemy) and defeats it ~80% of the time. The error backpropagation algorithm as applied to multi-layer perceptron and reinforcement learning is used for the robot’s attack or defence strategy. [Read More]