Uttaran Bhattacharya joined Adobe Research as a Research Scientist in September 2022. His research spans the areas of human motion understanding, computer vision, deep learning and generative modeling, and their applications in 2D and 3D motion synthesis, editing, video compression and video summarization.


Before joining Adobe Research, Uttaran completed his Ph.D. in Computer Science at the University of Maryland, College Park, where he was advised by Dr. Dinesh Manocha in the GAMMA Lab and was partially supported by an Adobe Research Fellowship. With a research focus on affective human motion detection and synthesis, he worked on automated techniques to detect emotions from 3D human body expressions such as gaits and gestures, and to generate animated 3D body expressions corresponding to different emotions in a variety of social contexts. He also interned at Adobe Research between 2020 and 2021, where he worked on novel problems in human-centric and user-specific video highlight detection.


More information about Uttaran can be found on his website.


HanDiffuser: Text-to-Image Generation With Realistic Hand Appearances

Narasimhaswamy, S., Bhattacharya, U., Chen, X., Dasgupta, I., Mitra, S., Hoai, M. (Jun. 19, 2024)

CVPR 2024

Speech2UnifiedExpressions: Synchronous Synthesis of Co-Speech Affective Face and Body Expressions from Affordable Inputs

Bhattacharya, U., Bera, A., Manocha, D. (Jun. 18, 2024)

CVPR 2024 HuMoGen Workshop

Show Me What I Like: Detecting User-Specific Video Highlights Using Content-Based Multi-Head Attention

Bhattacharya, U., Wu, G., Petrangeli, S., Swaminathan, V., Manocha, D. (Oct. 10, 2022)

ACM International Conference on Multimedia

HighlightMe: Detecting Highlights From Human-Centric Videos

Bhattacharya, U., Wu, G., Petrangeli, S., Swaminathan, V., Manocha, D. (Oct. 10, 2021)

IEEE/CVF International Conference on Computer Vision (ICCV)