Mission. I specialize in Scene Graph Generation (SGG) models, transcending traditional limitations through weakly supervised techniques. My research integrates knowledge graph-based and context-aware reasoning, empowering intelligent machines to comprehend the visual world deeply. Additionally, my work extends to refining vision-language models (VLMs), correlating information from visual and linguistic domains for nuanced understanding. With a focus on scene graphs and captioning techniques, I aim to enhance VLMs through instruction-tuning and synthetic data augmentation, contributing to the evolution of intelligent systems.

Availability of thesis topics. If you are not yet familiar with our thesis notes, I recommend you start here: Thesis Proposals at AIML
If you are interested in my research topic or have an exciting idea, feel free to reach out.