Ice Hockey Puck Localization Using Contextual Cues

2025-06-04

AI Summary

Hockey broadcast videos pose significant challenges for player detection due to small object sizes, frequent occlusions, motion blur, broadcast artifacts, and large scale variations. To address these issues, this work introduces player pose—specifically orientation and positioning—as a strong contextual cue and proposes a scale-adaptive single-frame precise localization framework. Methodologically, we design: (1) a context encoder integrating behavioral priors; (2) a multi-scale feature pyramid jointly optimized with a channel-gated decoder; (3) homography-based scale-invariant spatial mapping; and (4) the novel Rectified Spatial Localization Error (RSLE), the first perspective-bias-free metric for evaluating spatial localization accuracy on the hockey rink. Evaluated on PuckDataset, our method achieves a 12.2% improvement in mean Average Precision (mAP) and a 25% reduction in RSLE, establishing new state-of-the-art performance.

Technology Category

• Computer Vision: Motion & Tracking
• Intelligent Robots: State Estimation
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Location- and context-aware Web and WoT applications and services
• Security and Privacy: Large-scale security measurements

Abstract

Puck detection in ice hockey broadcast videos poses significant challenges due to the puck's small size, frequent occlusions, motion blur, broadcast artifacts, and scale inconsistencies due to varying camera zoom and broadcast camera viewpoints. Prior works focus on appearance-based or motion-based cues of the puck without explicitly modelling the cues derived from player behaviour. Players consistently turn their bodies and direct their gaze toward the puck. Motivated by this strong contextual cue, we propose Puck Localization Using Contextual Cues (PLUCC), a novel approach for scale-aware and context-driven single-frame puck detections. PLUCC consists of three components: (a) a contextual encoder, which utilizes player orientations and positioning as helpful priors; (b) a feature pyramid encoder, which extracts multiscale features from the dual encoders; and (c) a gating decoder that combines latent features with a channel gating mechanism. For evaluation, in addition to standard average precision, we propose Rink Space Localization Error (RSLE), a scale-invariant homography-based metric for removing perspective bias from rink space evaluation. The experimental results of PLUCC on the PuckDataset dataset demonstrated state-of-the-art detection performance, surpassing previous baseline methods by an average precision improvement of 12.2% and RSLE average precision of 25%. Our research demonstrates the critical role of contextual understanding in improving puck detection performance, with broad implications for automated sports analysis.

Problem

• Detecting small, occluded puck in hockey videos
• Leveraging player behavior cues for puck localization
• Improving scale-aware puck detection accuracy

Innovation

• Uses player orientations as contextual priors
• Employs feature pyramid for multiscale encoding
• Integrates gating decoder with channel mechanism