Every cricket fan has watched the Hawk-Eye animation play out on their television — the red ball curving through the air, hitting the pad, then the predicted trajectory extending toward the stumps in green, yellow, or red. Most fans understand that it helps make LBW decisions. Very few understand how it actually works. As someone who builds real-time 3D tracking and simulation systems for a living, I find the engineering behind Hawk-Eye genuinely impressive — and worth explaining properly.
This is a breakdown of how Hawk-Eye and the full DRS system works technically, what AI has added to it recently, and why the IPL has been leading the world in how this technology is being applied.
What Hawk-Eye Actually Is
Hawk-Eye is a computer vision and ball-tracking system developed by British engineers and first used in cricket in the 2001 Ashes series. At its core it is a multi-camera triangulation system — six cameras placed around the ground capture the ball from different angles simultaneously, and a central processing system combines those feeds to reconstruct the ball's three-dimensional position in real time.
From a developer's perspective, this is essentially the same problem as real-time 3D object tracking — something I work with regularly in AR and XR applications. The challenge in cricket is speed and precision: a fast ball travels at over 140km/h, the tracking window between release and impact is fractions of a second, and the margin for error in an LBW decision is measured in millimetres. The engineering tolerances are demanding.
The system does not just track where the ball went — it predicts where it would have gone if the batsman had not been there. That predictive trajectory is what you see on screen during an LBW review. Hawk-Eye builds a physics model of the ball's behaviour based on its observed trajectory up to the point of impact, then extends that model forward to the stumps.
The Full DRS System — More Than Just Hawk-Eye
Hawk-Eye is one component of a broader Decision Review System. The full DRS stack includes several distinct technologies working together:
UltraEdge (formerly Snickometer) — detects sound vibrations using a microphone in the stumps and correlates them with the video frame. When a ball passes the edge of the bat, it creates a characteristic vibration signature. UltraEdge shows this as a waveform spike on the broadcast screen — the spike appearing at exactly the moment the ball passes the bat is strong evidence of a nick.
HotSpot — uses infrared cameras to detect heat generated by friction when ball contacts bat or pad. A bright spot appears on the infrared image at the point of contact. HotSpot is the most conclusive technology for edge detection but is expensive and not used in all tournaments.
Ball Tracking (Hawk-Eye) — the trajectory system described above, used primarily for LBW decisions and now increasingly for no-ball and wide decisions in the IPL.
Smart Replay System — introduced in IPL 2024, this automates the switching between camera angles during reviews, bringing up the most relevant view immediately without requiring manual camera selection. It significantly speeds up the review process.
What AI Has Added in 2025-2026
The most significant recent development is AI-enhanced ball detection within Hawk-Eye. Traditional computer vision ball tracking can struggle with certain conditions — ball colour blending with pitch, motion blur at high speeds, occlusion by fielders. AI-backed detection, trained on millions of frames of cricket footage, significantly reduces these errors.
In 2025, Hawk-Eye's AI was enhanced to differentiate between pitch types and bounce behaviours. A ball pitching on a damp pitch behaves differently from one on a dry, cracked surface — the trajectory model now accounts for this dynamically rather than using a static physics model. For an Indian cricket context where pitches vary enormously across venues and conditions, this is a meaningful improvement in accuracy.
The IPL has been at the forefront of applying this technology. In IPL 2024, ball-tracking was used for the first time to judge over-the-waist no-balls. In IPL 2025, Hawk-Eye was extended to judge head-high wides — measuring the ball's height as it passes the batsman at the popping crease and comparing it to that batsman's pre-recorded head height. In IPL 2026, the Smart Replay System continued to streamline reviews further. The IPL's willingness to adopt technology before other cricket boards is one of the reasons it consistently leads global cricket in production quality.
From an XR Developer's Perspective
What strikes me about Hawk-Eye from my work in XR is the real-time 3D rendering pipeline. The visualisation you see on television — the ball's trajectory rendered as a smooth 3D animation with the stumps and ground plane accurately represented — is essentially a real-time XR overlay on a real-world cricket ground. The coordinate system has to be calibrated precisely to the physical ground layout, the camera positions, and the stump positions before each match.
This is the same calibration challenge we face in AR applications — aligning a virtual coordinate system with a real-world physical space precisely enough that the overlay looks correct. In AR apps I build for enterprise clients, centimetre-level accuracy is often sufficient. For Hawk-Eye, the tolerances are millimetre-level because an LBW decision may hinge on whether the predicted ball trajectory passes inside or outside the edge of the stump. The engineering required to achieve that in a live broadcast environment, across different grounds, weather conditions, and camera setups, is genuinely impressive.
The Ongoing Debates
Hawk-Eye is not without controversy. The most persistent debate is around Umpire's Call — the rule that if the ball is predicted to hit the stumps but only marginally (less than 50% of the ball overlapping), the on-field decision stands regardless of what the tracking shows. Critics argue this undermines the purpose of the review system. Defenders argue it acknowledges the inherent uncertainty in any predictive model — the trajectory is a probability, not a certainty, and Umpire's Call represents that uncertainty honestly.
From a technical standpoint, both sides have merit. Hawk-Eye's trajectory prediction is accurate within its stated tolerances, but those tolerances do exist. A ball predicted to hit the top of off stump by 2mm may or may not actually have done so. The system is very good — but it is not infallible, and Umpire's Call is an honest acknowledgement of that.
What Comes Next
The direction of travel is clear: more automation, less human judgment in routine decisions. Automatic no-ball calling is already deployed in some tournaments. AI-assisted wide calling is expanding. The logical endpoint — fully automated first-instance decisions without requiring a review — is technically achievable now. The question is whether cricket's governing bodies want to go there, and how fans and players will respond to removing the human umpire from more decisions.
For someone who builds AI-integrated XR systems, watching cricket technology evolve is watching the same principles I apply in industrial training environments being applied to sport at broadcast scale. The underlying engineering — computer vision, real-time 3D tracking, predictive physics models, AI-enhanced detection — is the same. Cricket just happens to be doing it in front of hundreds of millions of viewers.
Frequently Asked Questions
Hawk-Eye's stated accuracy is within 2.6mm for ball-tracking. That is extremely precise but not perfect — which is why Umpire's Call exists for marginal decisions. The AI enhancements in 2025 have further reduced errors in difficult conditions like high ball speeds and varied pitch surfaces.
Six dedicated tracking cameras placed around the ground, plus additional broadcast cameras that feed into the system. The six tracking cameras are specifically positioned and calibrated to the ground layout before each match to ensure the 3D coordinate system is accurate.
Introduced in IPL 2024, the Smart Replay System automates camera angle switching during DRS reviews, bringing up the most relevant angles automatically without manual selection. It reduces review time and ensures the third umpire sees the best available evidence immediately. IPL 2025 and 2026 have continued using it.
Hawk-Eye is an augmented reality broadcast overlay — it adds computer-generated 3D graphics onto real-world cricket footage in real time. The same fundamental technologies used in AR applications (computer vision, 3D spatial mapping, real-time rendering) underpin how the Hawk-Eye visualisation is produced and calibrated to each ground.