Adithya N.

About Adithya N.

I’m a Master’s student in Computer Vision at Carnegie Mellon University, where I work in the Human Sensing Lab under Prof. Fernando De la Torre. My research focuses on extending 2D Vision-Language Models (VLMs) toward zero-shot 3D understanding. I also work on control for video diffusion models.I’m broadly interested in how multimodal models perceive and represent the world — particularly how symbolic and visual reasoning emerge within large-scale vision systems. My work spans generative modeling, 3D perception, and scene understanding.Previously, I’ve worked on projects at the intersection of computer vision and graphics, including differentiable rendering for adversarial scene exploration, and medical imaging pipelines that improved clinical measurement reliability.I’m passionate about designing AI systems that learn from perception, adapt from context, and generalize beyond their training data — bringing us closer to truly grounded, multimodal intelligence.

Experience

1. Research Engineer

   HeyGen

   Jan 2026 — Present · Los Angeles, CA, US

   Joint audio+video generation with GRMs to improve naturalness- Better audio generation models to improve ambient sounds, speech prosody and vocal cloning- Async ray pipelines +vLLM for serving VLMs at scale

Education

• Tenby Schools Ipoh

  High School Diploma

• Carnegie Mellon University

  Master of Science - MS, Computer Vision