Antareep Dey

I'm a Student Researcher at IIT Kharagpur, working on computer vision with a focus on video understanding, test-time adaptation, and vision-language multimodality. I'm especially interested in the capabilities that emerge from the unique interactions between vision and language, capabilities neither modality can achieve on its own, and understanding what drives that emergence is what keeps me curious. Previously, I've also worked on explainable AI for medical imaging, small language models, and edge-deployed vision systems.

In my free time, you'll find me obsessing over Formula 1 🏎️ ,marvelling at architecture 🏛️, or feeding my inner shutterbug📸. I firmly believe the best ideas come from conversations, preferably caffeinated ones ☕. If you're interested in my research ( and even if you aren't ) let's connect! 💬