About nitish
I’m a results-driven Computer Vision and Machine Learning specialist recognized for delivering groundbreaking solutions across industries—ranging from sports and physical security to gaming. Backed by patented innovations and peer-reviewed research in CV / ML, I excel at transforming complex high-level concepts into real-world applications such as markerless motion capture, multi-view multi-object tracking, 3D reconstruction, and advanced analytics. I’m passionate about mentoring teams, leveraging AI to tackle the toughest challenges, and constantly pushing the boundaries of what’s possible.
Key Skills
Experience
Lead Computer Vision Engineer
CurrentKinatrax
On-Field Player Tracking in 3D: Championed the development of State-Of-The-Art tracking system for players in 3D for professional sports, employing YOLOX for detection, K-Partite Graph matching employing a novel methodology of Hungarian Algorithm invoked recursively for cross-view aggregation, and Kalman Filter for tracking. Collaborated closely with stakeholders ensuring the patent pending solution’s implementation across NBA and MLB stadiums nationwide. Baseball Centroid and Spin Tracking: Served as a key mentor within the computer vision team, building baseball tracking capabilities that accurately measure ball centroid, spin rate, and direction. Pioneered a framework that uses DeepLab V3+ to estimate ball seam segmentation masks, integrating subsequent layers to estimate 3D orientation via quaternions and eliminate gimbal lock issues. Proposed innovative methods to accurately estimate spin despite symmetrical seams. Automated Camera Drift Correction: Spearheaded the development of an automated framework utilizing LightGlue as a keypoint matching algorithm to estimate camera drift over time in order to accurately infer 3D positions of tracked objects using the camera parameters. 3D Reconstruction and Novel View Synthesis: Early in the initiative of prototyping NeRF and Gaussian-Splatting based pipelines that fuse synchronized multi-camera footage from stadiums and courts into metrically accurate, photorealistic 3D scene assets.
Senior Computer Vision Engineer
Verkada
Machine Learning Engineer
Visual Concepts
* Researched and developed pose estimation, character transition networks, and 3D mesh generation for motion prediction and motion matching to improve the animation visual quality and optimize production workflows for the NBA 2K video game. * Implemented multi-view players and ball tracking to identify appropriate regions to crop for player pose estimation using real game footage videos. * Implemented 0-9 digit detection in the quarter timer to accurately map quarter time to valid frame numbers helping identify valid frames with a specified action sequence.
Education
New York University
Master Of Science
Indian Institute Of Technology, Delhi
Bachelor Of Technology
New York University
Interested in connecting with nitish?
Sign up for NinjaHire to send a connection request.
Common Questions
What is nitish's expertise?
nitish specializes in Computer Vision Engineer, with expertise in artificial intelligence, c (programming language), c++, computer vision, convolutional neural networks.
Where is nitish located?
nitish is based in san mateo, california, united states.
How much experience does nitish have?
nitish has 11+ years of professional experience.
How can I contact nitish?
You can connect with nitish through NinjaHire by signing up for a free account.