Video-only camera pose estimation for sports broadcast graphics using venue-aware deep learning
Figure 4 – Overlay: Projected model + Detected lines

By F. Maurício, wTVision 12 September 2026

Tech Papers 2026: This paper presents a video-only camera pose estimation framework that eliminates the need for dedicated tracking hardware.

Abstract 

Sports broadcast graphics rely on accurate camera tracking to enable spatially aligned visual augmentation. Traditional approaches depend on hardware-based tracking systems or manual calibration workflows, increasing operational complexity and deployment cost.

You are not signed in

Only registered users can read the rest of this article.

Favourites:

Registered users only: Login