Jerrin Bright

I am a Founding Engineer at Mecka AI, building the data layer for humanoids. We collect large-scale real-world data to train emboddied agents, where I drive 3D human demonstration understanding which feeds into Vision-Language Action (VLA) and world model pipelines, serving pioneering robotics and AI companies. I'm also a full-time doctoral student at the Vision and Image Processing Lab, University of Waterloo, supervised by Dr. John Zelek.

My research focuses on bridging the gap between human and machine perception, exploring how artificial systems can achieve robust world understanding from sparse visual inputs. I draw inspiration from the way humans rapidly infer geometry, motion, and dynamics using limited observations. My broader interests include human motion understanding, vision-language-action modeling, and large-scale multimodal data systems, viewing data itself as an indispensable layer in the development of next-gen intelligent, interactive agents.

Previously, I collaborated with the Baltimore Orioles (MLB) on robust baseball player kinematics analysis and performance evaluation. During my undergrad, I worked with McMaster University, Arizona State University, Yuan-Ze University, and Indian Institute of Science on UAV navigation, visual odometry, camera calibration, and robotics.

Beyond academia, I am an avid cricket and chess enthusiast who enjoys travelling, kick-boxing, skating, sketching, and working out at the gym.

Email  |  Scholar  |  X  |  LinkedIn

profile photo
Mecka UWaterloo Orioles IISc
McMaster ASU Yuan-Ze VIT

News

  • [Jan 2025] DreamPose3D submitted to IEEE Transactions on Multimedia
  • [Dec 2025] Attended NeurIPS 2025 in San Diego
  • [Sep 2025] Joined Mecka AI as Founding Engineer
  • [Aug 2025] On-site with Baltimore Orioles for PitcherNet deployment
  • [Jun 2025] Attended CVPR 2025 in Nashville
  • [May 2025] Gen4D accepted to CVPR Workshop 2025
  • [May 2025] SportMamba accepted as Oral at CVPR Workshop 2025
  • [April 2025] Featured in Toronto Star, CBC, CTV, Globe and Mail for PitcherNet
  • [Sep 2024] TokenCliPose accepted to NeurIPS 2024
  • [May 2024] Best Paper Award at CRV 2024
  • [Apr 2024] PitcherNet accepted as Oral at CVPR Workshop 2024

Current Focus

At Mecka AI: Building scalable pipelines for 3D human demonstration understanding from egocentric data (hand/body pose estimation), to power next-gen emboddied AI agents training.

PhD Research: Investigating 4D world reconstruction from strong human prior modeling and robust camera-scene-human joint reasoning from monocular observations.

@article{bright2025avatar4d, title={Avatar4D: Synthesizing Domain-Specific 4D Humans for Real-World Pose Estimation}, author={Bright, Jerrin and Wang, Zhibo and Klepachevskyi, Dmytro and Chen, Yuhao and Rambhatla, Sirisha and Clausi, David and Zelek, John}, journal={arXiv preprint arXiv:2512.16199}, year={2025} }
@article{bright2025dreampose3d, title={DreamPose3D: Hallucinative Diffusion with Prompt Learning for 3D Human Pose Estimation}, author={Bright, Jerrin and Chen, Yuhao and Zelek, John}, journal={arXiv preprint arXiv:2511.09502}, year={2025} }
@inproceedings{iaboni2025position, title={Position Paper: New Views of Shots - Towards Measures of Net Visibility and Reachability}, author={Iaboni, Evan and Negulescu, Sebastian and Pitassi, Miles and Nazemi, Amir and Bright, Jerrin and Chomko, Vasyl and Clausi, David A and Dickinson, Sven and Brecht, Tim}, booktitle={Linkoping Hockey Analytics Conference}, year={2025} }
@inproceedings{bright2025gen4d, title={Gen4D: Synthesizing Humans and Scenes in the Wild}, author={Bright, Jerrin and Wang, Zhibo and Chen, Yuhao and Rambhatla, Sirisha and Clausi, David and Zelek, John}, booktitle={IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops}, year={2025} }
@inproceedings{khanna2025sportmamba, title={SportMamba: Adaptive Non-Linear Multi-Object Tracking with State Space Models for Team Sports}, author={Khanna, Dheeraj and Bright, Jerrin and Chen, Yuhao and Zelek, John}, booktitle={IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops}, year={2025} }
@inproceedings{balaji2024seeing, title={Seeing Beyond the Crop: Using Language Priors for Out-of-Bounding Box Keypoint Prediction}, author={Balaji, Bavesh and Bright, Jerrin and Chen, Yuhao and Rambhatla, Sirisha and Zelek, John and Clausi, David}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems (NeurIPS)}, year={2024} }
@inproceedings{bright2024pitchernet, title={PitcherNet: Powering the Moneyball Evolution in Baseball Video Analytics}, author={Bright, Jerrin and Balaji, Bavesh and Chen, Yuhao and Clausi, David A and Zelek, John}, booktitle={IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops}, year={2024} }
@inproceedings{bright2024d2ahmr, title={Distribution and Depth-Aware Transformers for 3D Human Mesh Recovery}, author={Bright, Jerrin and Balaji, Bavesh and Prakash, Harish and Chen, Yuhao and Clausi, David A and Zelek, John}, booktitle={21st Conference on Robots and Vision (CRV)}, year={2024} }
@inproceedings{balaji2024domain, title={Domain-Guided Masked Autoencoders for Unique Player Identification}, author={Balaji, Bavesh and Bright, Jerrin and Rambhatla, Sirisha and Chen, Yuhao and Wong, Alexander and Zelek, John and Clausi, David A}, booktitle={21st Conference on Robots and Vision (CRV)}, year={2024} }
@inproceedings{bright2023mitigating, title={Mitigating Motion Blur for Robust 3D Baseball Player Pose Modeling for Pitch Analysis}, author={Bright, Jerrin and Chen, Yuhao and Zelek, John}, booktitle={Proceedings of the 6th International Workshop on Multimedia Content Analysis in Sports (MMSports)}, pages={63--71}, year={2023}, organization={ACM} }
@phdthesis{bright2025worldreconstruction, title={4D World Reconstruction of Humans, Scenes, and Camera Systems}, author={Bright, Jerrin}, school={University of Waterloo}, year={2025}, note={PhD Comprehensive Exam} }
@mastersthesis{bright2024robust, title={Robust 3D Human Modeling for Baseball Sports Analytics}, author={Bright, Jerrin}, school={University of Waterloo}, year={2024} }
@bachelorsthesis{bright2022uav, title={An End-to-End Autonomous UAV System in GPS-Denied and Unstructured Environments}, author={Bright, Jerrin and Rajkumar, Suryaprakash and Selvakumar, Arockia}, school={Vellore Institute of Technology}, year={2022} }

Publications

Representative papers are highlighted. See Google Scholar for a full list.

Avatar4D
Avatar4D: Synthesizing Domain-Specific 4D Humans for Real-World Pose Estimation
Jerrin Bright, Zhibo Wang, Dmytro Klepachevskyi, Yuhao Chen, Sirisha Rambhatla, David Clausi, John Zelek
arXiv preprint, 2025
ArXiv | BibTeX

Synthesizing domain-specific 4D humans for improving real-world pose estimation.
DreamPose3D
DreamPose3D: Hallucinative Diffusion with Prompt Learning for 3D Human Pose Estimation
Jerrin Bright, Yuhao Chen, John Zelek
Submitted to IEEE Transactions on Multimedia, 2026
ArXiv | BibTeX

A hallucinative diffusion approach with prompt learning for 3D human pose estimation.
Hockey Analytics
Position Paper: New Views of Shots - Towards Measures of Net Visibility and Reachability
Evan Iaboni, Sebastian Negulescu, Miles Pitassi, Amir Nazemi, Jerrin Bright, Vasyl Chomko, David A. Clausi, Sven Dickinson, Tim Brecht
Linkoping Hockey Analytics Conference, 2025
Paper | BibTeX

Novel measures for analyzing shot quality through net visibility and reachability in hockey.
Gen4D
Gen4D: Synthesizing Humans and Scenes in the Wild
Jerrin Bright, Zhibo Wang, Yuhao Chen, Sirisha Rambhatla, David Clausi, John Zelek
CVPR Workshop, 2025  Poster
Project Page | ArXiv | Poster | BibTeX

A framework for synthesizing realistic 4D humans and scenes in unconstrained environments.
SportMamba
SportMamba: Adaptive Non-Linear Multi-Object Tracking with State Space Models for Team Sports
Dheeraj Khanna, Jerrin Bright, Yuhao Chen, John Zelek
CVPR Workshop, 2025
Paper | Poster | BibTeX

Adaptive multi-object tracking using state space models for team sports applications.
NeurIPS
Seeing Beyond the Crop: Using Language Priors for Out-of-Bounding Box Keypoint Prediction
Bavesh Balaji, Jerrin Bright, Yuhao Chen, Sirisha Rambhatla, John Zelek, David Clausi
NeurIPS, 2024  Poster
Paper | OpenReview | Slides | Poster | BibTeX

Leveraging language priors to predict keypoints that fall outside the bounding box crop.
PitcherNet
PitcherNet: Powering the Moneyball Evolution in Baseball Video Analytics
Jerrin Bright, Bavesh Balaji, Yuhao Chen, David A Clausi, John Zelek
CVPR Workshop, 2024  Oral
ArXiv | Paper | Video | Slides | Poster | BibTeX

A robust framework for baseball pitcher biomechanics analysis from broadcast videos.
D2AHMR
Distribution and Depth-Aware Transformers for 3D Human Mesh Recovery
Jerrin Bright, Bavesh Balaji, Harish Prakash, Yuhao Chen, David A Clausi, John Zelek
Conference on Robots and Vision (CRV), 2024  Oral  Best Paper Award
ArXiv | Paper | Video | Slides | Poster | BibTeX

A transformer-based approach for 3D human mesh recovery with distribution and depth awareness.
Domain-Guided MAE
Domain-Guided Masked Autoencoders for Unique Player Identification
Bavesh Balaji, Jerrin Bright, Sirisha Rambhatla, Yuhao Chen, Alexander Wong, John Zelek, David A Clausi
Conference on Robots and Vision (CRV), 2024  Oral
ArXiv | Paper | Slides | Poster | BibTeX

Domain-guided masked autoencoders for unique player identification in sports videos.
Motion Blur
Mitigating Motion Blur for Robust 3D Baseball Player Pose Modeling for Pitch Analysis
Jerrin Bright, Yuhao Chen, John Zelek
ACM MMSports Workshop, 2023
Paper | Slides | Poster | BibTeX

Addressing motion blur challenges for robust 3D pose estimation of baseball players during pitch analysis.

Thesis
PhD Comprehensive Exam
4D World Reconstruction of Humans, Scenes, and Camera Systems
Jerrin Bright, John Zelek
PhD Comprehensive Exam, University of Waterloo, 2026
PDF | Slides | BibTeX
thesis
Robust 3D Human Modeling for Baseball Sports Analytics
Jerrin Bright, John Zelek
Master's Thesis, University of Waterloo, 2024
UWSpace | PDF | Slides | BibTeX
thesis
An End-to-End Autonomous UAV System in GPS-Denied and Unstructured Environments
Jerrin Bright, Suryaprakash Rajkumar, Arockia Selvakumar
Bachelor's Thesis, VIT Chennai, 2022
PDF | Slides | BibTeX

Awards and Honors

Scholarships:

  • International Doctoral Student Award, University of Waterloo
  • MITACS Accelerate International Award, University of Waterloo
  • International Master's Award of Excellence, University of Waterloo
  • Graduate Research Studentship, University of Waterloo
  • MITACS Graduate Research Fellowship, University of Waterloo
  • MITACS Globalink Graduate Fellowship, MITACS Canada

Honors and Awards:

  • Best Paper Award (Computer Vision), Conference on Robots and Vision, 2024
  • Outstanding Paper Award, RIACT International Conference, 2020
  • Best Outgoing Student - Robotics Club, Vellore Institute of Technology, Chennai, India
  • Best Club Award - Robotics Club, University Day 2021, Vellore Institute of Technology, Chennai, India
  • Co-Founder and Captain, Atom Robotics - An intelligent robotics and space exploration team @ VIT University, India
  • Winner in 15+ Robotics Competitions, Held in various Techfests in universities around India

Academic Service

Reviewer: CVPR, NeurIPS, 3DV, IROS, EAAI, CVSports, CVIU, MMSports, TNNLS, TCSVT, CVIS, CTIS, IMAVIS, The Journal of Supercomputing

Technical Program Member: ACM MMSports @ ACM Multimedia 2024

Teaching Assistant:

  • SYDE461: Systems Design Capstone Project 1 (Fall 2023, Fall 2024)
  • SYDE462: Systems Design Capstone Project 2 (Winter 2024, Winter 2025)
  • SYDE361: Systems Design Methods 1: Needs Analysis and Prototyping (Summer 2024)
  • BME361: Biomedical Engineering Design (Winter 2023)
  • BME101L: Communications in Biomedical Engineering: Visualization (Fall 2022)

Media Coverage

My work has been featured in:

  • Toronto Star: "The Baltimore Orioles are developing a game-changing baseball technology, with help from a Canadian university" (April 18, 2025)
  • Baltimore Sun: "Orioles developing form of AI technology that could someday change scouting forever" (April 23, 2025)
  • UWaterloo Media: "PitcherNet helps researchers throw strikes with AI analysis" (April 23, 2025)
  • CTV News: "Making a pitch to change the game: Waterloo tech to analyze baseball pitcher's performance with smartphone" (April 29, 2025)
  • CBC News: "UW researchers use AI to analyze baseball players' movements" (April 30, 2025)
  • The Globe and Mail: "University of Waterloo, Orioles partner on AI project to track pitcher biomechanics" (May 6, 2025)
  • Global News: "University of Waterloo researchers develop AI technology for Baltimore Orioles" (May 13, 2025)
  • Baseball HQ Radio: "Friday Full Edition w Jerrin Bright" (May 23, 2025)

Design inspired by Jon Barron. Last updated 2026.