Medhini Narasimhan

Hi! I'm a Research Scientist at Google Deepmind working on Generative AI, specifically Veo, our video generation model. I work on the core model design and training, and also on Veo's control capabilities. I developed the Image-to-Video feature for Veo that was showcased at Google IO 2024. I also lead the effort to productionize Veo in VideoFX, YouTube DreamScreen, and Google Cloud's Vertex AI.

I graduated with a PhD in Computer Science from UC Berkeley in 2023, where I was advised by Prof. Trevor Darrell and a member of Berkeley AI Research (BAIR). My thesis focused on developing large multimodal models capable of learning representations from videos, specifically to create short visual summaries.

Prior to Berkeley, I completed my Master's in Computer Science at the UIUC advised by Prof. A. Schwing and Prof. S. Lazebnik. In 2017, I obtained my Bachelor's Degree from NITK, India.

I have been fortunate to intern at Facebook AI Research in Summer 2019 and 2022, Google Research in Summer 2021 and at Zillow Research in Summer 2018.

I am a honored to be a Snap Research Scholar (2019) and a Siebel Scholar (2018)!

Email |  CV |  Google Scholar |  Github |  LinkedIn

Research
Learning and Verification of Task Structure from Instructional Videos
 
Medhini Narasimhan, Licheng Yu, Sean Bell, Ning Zhang, Trevor Darrell
arXiv 2023 (In Submission)
 
Paper | Website | Code
TL;DW? Summarizing Instructional Videos with Task Relevance & Cross-Modal Saliency
 
Medhini Narasimhan, Arsha Nagrani, Chen Sun, Miki Rubinstein, Trevor Darrell, Anna Rohrbach, Cordelia Schmid
ECCV 2022
 
Paper | Website | Code
CLIP-It! Language-Guided Video Summarization
 
Medhini Narasimhan, Anna Rohrbach, Trevor Darrell
Neurips 2021
 
Paper | Website | Code
Strumming to the Beat: Audio-Conditioned Contrastive Video Textures
(Best Paper, Honorable Mention)
 
Medhini Narasimhan, Shiry Ginosar, Andrew Owens, Alyosha Efros,
Trevor Darrell
WaCV 2022
 
Paper | Website | Code | Video | AICC CVPR Workshop Paper
Seeing the Un-Scene: Learning Amodal Semantic Maps for Room Navigation
 
Medhini Narasimhan, Erik Wijmans, Xinlei Chen, Trevor Darrell, Dhruv Batra,
Devi Parikh, Amanpreet Singh
ECCV 2020
 
Paper
Out of the Box: Reasoning with Graph Convolution Nets for Factual Visual Question Answering
 
Medhini Narasimhan, Svetlana Lazebnik, Alexander Schwing
Neurips 2018
 
Paper | Poster | Video | Bib
Straight to the Facts: Learning Knowledge Base Retrieval for Factual Visual Question Answering
 
Medhini Narasimhan, Alexander Schwing
ECCV 2018
 
Paper | Poster | Bib
Dynamic video anomaly detection and localization using sparse denoising autoencoders
 
Medhini Narasimhan, Sowmya Kamath
Multimedia Tools and Applications Journal (MTAP), 2017
 
Paper | Code | Bib

Image Credits: Link

EGA-FMC: Enhanced Genetic Algorithm based Fuzzy K-Modes Clustering for Categorical Data
 
Medhini Narasimhan, Balaji Balasubramanian, Suryansh Kumar, Nagamma Patil
International Journal of Bio-Inspired Computation, 2018
 
Paper | Code | Bib
Predicting Symptom Severity and Contagiousness of Respiratory Viral Infections (Best Poster Award)
 
Medhini Narasimhan, Guiseppe Vietri, Arpit Mehta, Farid Rajabli, Vanessa Aguiar-Pulido, Kalai Mathee, Giri Narasimhan
ISMB 2016
 
[Poster]
Teaching

I have been a teaching assistant for the following courses:

Community Service

I have reviewed for:

  • Conferences: CVPR, ICCV, ECCV, NeurIPS, CoRL
  • Journals: PAMI, CVIU


Cloned from here!