Team

My mission is to advance the field of scene understanding and reasoning, pushing the boundaries of visual intelligence. I aim to develop Scene Graph Generation (SGG) models that are closer to reality by transcending traditional limitations and reducing annotation dependency through weakly supervised and self-supervised techniques. My research also focuses on creating reasoning based SGGs with the integration of advanced reasoning mechanisms into these models by incorporating knowledge graph-based, neuro-symbolic and context-aware reasoning capabilities. Ultimately, my aim is to empower intelligent machines and augmented realities with a deeper comprehension of the visual world, uncovering its hidden mysteries.