Multi Player Tracking in Ice Hockey with Homographic Projections
0
⛏️
Sign in to get full access
Overview
- Tackles the problem of multi-object tracking (MOT) in ice hockey broadcasts
- Aims to accurately localize and associate players across a video sequence
- Addresses challenges such as occlusions, blurs, and agile player movements common in broadcast footage
Plain English Explanation
The paper focuses on a computer vision problem called multi-object tracking (MOT) in ice hockey broadcasts. The goal is to track individual players and maintain their identities as they move around the rink during a game. This is an important task that can enable various analytics and enhance the viewing experience for audiences.
However, existing tracking approaches struggle with the unique challenges of ice hockey footage, such as players occluding each other, blurred images, and the fast-paced, agile movements of the players. To address these issues, the researchers propose a novel tracking method that frames MOT as a bipartite graph matching problem combined with homography.
The key idea is to disentangle the positions of occluded and overlapping players by mapping their foot keypoints to an overhead view of the rink. This projected spatial information is then encoded into the graph network to provide reliable context for consistent player tracking and uninterrupted tracklet prediction.
The results show significant improvements in two important MOT metrics, IDsw and IDF1, on two available datasets of ice hockey broadcast footage.
Technical Explanation
The paper formulates the MOT problem in ice hockey as a bipartite graph matching task, with the addition of homography to address the challenges of occlusions and overlapping players common in broadcast video.
The proposed approach first extracts player foot keypoints from the broadcast view and maps them to an overhead rink template using homography. This disentangles the positions of occluded and overlapping players, providing reliable spatial context for the tracking task.
The projected player positions are then encoded into a graph network, where nodes represent player detections and edges represent potential associations. The graph is optimized using a bipartite matching algorithm to generate consistent player tracklets across the video sequence.
The authors evaluate their method on two ice hockey datasets and report significant improvements in both the IDsw and IDF1 metrics compared to existing MOT approaches. This demonstrates the effectiveness of their homography-based spatial encoding for robust and accurate player tracking in complex broadcast scenarios.
Critical Analysis
The paper presents a novel and promising approach to addressing the challenges of MOT in ice hockey broadcasts. The use of homography to disentangle the positions of occluded and overlapping players is a clever solution that provides valuable spatial context for the tracking task.
However, the paper does not provide a detailed analysis of the limitations or potential drawbacks of the proposed method. For instance, the performance of the homography estimation and its sensitivity to factors like camera calibration or player occlusions could be further investigated.
Additionally, the paper focuses solely on ice hockey, and it would be interesting to see how the approach generalizes to other sports with similar challenges, such as soccer or basketball. Exploring the transferability of the method to different domains could further validate its broader applicability.
Overall, the research presents a compelling solution to a relevant computer vision problem and demonstrates the potential of leveraging homography for robust multi-object tracking in complex, dynamic environments.
Conclusion
This paper introduces a novel approach to multi-object tracking in ice hockey broadcasts by formulating the problem as a bipartite graph matching task combined with homography. The key innovation is the use of homography to disentangle the positions of occluded and overlapping players, providing reliable spatial context for consistent player tracking and uninterrupted tracklet prediction.
The results show significant improvements in two important MOT metrics, highlighting the effectiveness of the proposed method in dealing with the unique challenges of ice hockey broadcast footage. While the paper focuses on a specific sport, the underlying principles could potentially be applied to enhance multi-object tracking in other dynamic, occlusion-prone environments, contributing to the advancement of computer vision and video analytics.
This summary was produced with help from an AI and may contain inaccuracies - check out the links to read the original source documents!
Related Papers
⛏️
0
Multi Player Tracking in Ice Hockey with Homographic Projections
Harish Prakash, Jia Cheng Shang, Ken M. Nsiempba, Yuhao Chen, David A. Clausi, John S. Zelek
Multi Object Tracking (MOT) in ice hockey pursues the combined task of localizing and associating players across a given sequence to maintain their identities. Tracking players from monocular broadcast feeds is an important computer vision problem offering various downstream analytics and enhanced viewership experience. However, existing trackers encounter significant difficulties in dealing with occlusions, blurs, and agile player movements prevalent in telecast feeds. In this work, we propose a novel tracking approach by formulating MOT as a bipartite graph matching problem infused with homography. We disentangle the positional representations of occluded and overlapping players in broadcast view, by mapping their foot keypoints to an overhead rink template, and encode these projected positions into the graph network. This ensures reliable spatial context for consistent player tracking and unfragmented tracklet prediction. Our results show considerable improvements in both the IDsw and IDF1 metrics on the two available broadcast ice hockey datasets.
Read more5/24/2024
0
View-Centric Multi-Object Tracking with Homographic Matching in Moving UAV
Deyi Ji, Siqi Gao, Lanyun Zhu, Qi Zhu, Yiru Zhao, Peng Xu, Hongtao Lu, Feng Zhao, Jieping Ye
In this paper, we address the challenge of multi-object tracking (MOT) in moving Unmanned Aerial Vehicle (UAV) scenarios, where irregular flight trajectories, such as hovering, turning left/right, and moving up/down, lead to significantly greater complexity compared to fixed-camera MOT. Specifically, changes in the scene background not only render traditional frame-to-frame object IOU association methods ineffective but also introduce significant view shifts in the objects, which complicates tracking. To overcome these issues, we propose a novel universal HomView-MOT framework, which for the first time, harnesses the view Homography inherent in changing scenes to solve MOT challenges in moving environments, incorporating Homographic Matching and View-Centric concepts. We introduce a Fast Homography Estimation (FHE) algorithm for rapid computation of Homography matrices between video frames, enabling object View-Centric ID Learning (VCIL) and leveraging multi-view Homography to learn cross-view ID features. Concurrently, our Homographic Matching Filter (HMF) maps object bounding boxes from different frames onto a common view plane for a more realistic physical IOU association. Extensive experiments have proven that these innovations allow HomView-MOT to achieve state-of-the-art performance on prominent UAV MOT datasets VisDrone and UAVDT.
Read more5/15/2024
0
Track Initialization and Re-Identification for~3D Multi-View Multi-Object Tracking
Linh Van Ma, Tran Thien Dat Nguyen, Ba-Ngu Vo, Hyunsung Jang, Moongu Jeon
We propose a 3D multi-object tracking (MOT) solution using only 2D detections from monocular cameras, which automatically initiates/terminates tracks as well as resolves track appearance-reappearance and occlusions. Moreover, this approach does not require detector retraining when cameras are reconfigured but only the camera matrices of reconfigured cameras need to be updated. Our approach is based on a Bayesian multi-object formulation that integrates track initiation/termination, re-identification, occlusion handling, and data association into a single Bayes filtering recursion. However, the exact filter that utilizes all these functionalities is numerically intractable due to the exponentially growing number of terms in the (multi-object) filtering density, while existing approximations trade-off some of these functionalities for speed. To this end, we develop a more efficient approximation suitable for online MOT by incorporating object features and kinematics into the measurement model, which improves data association and subsequently reduces the number of terms. Specifically, we exploit the 2D detections and extracted features from multiple cameras to provide a better approximation of the multi-object filtering density to realize the track initiation/termination and re-identification functionalities. Further, incorporating a tractable geometric occlusion model based on 2D projections of 3D objects on the camera planes realizes the occlusion handling functionality of the filter. Evaluation of the proposed solution on challenging datasets demonstrates significant improvements and robustness when camera configurations change on-the-fly, compared to existing multi-view MOT solutions. The source code is publicly available at https://github.com/linh-gist/mv-glmb-ab.
Read more5/30/2024
0
Basketball-SORT: An Association Method for Complex Multi-object Occlusion Problems in Basketball Multi-object Tracking
Qingrui Hu, Atom Scott, Calvin Yeung, Keisuke Fujii
Recent deep learning-based object detection approaches have led to significant progress in multi-object tracking (MOT) algorithms. The current MOT methods mainly focus on pedestrian or vehicle scenes, but basketball sports scenes are usually accompanied by three or more object occlusion problems with similar appearances and high-intensity complex motions, which we call complex multi-object occlusion (CMOO). Here, we propose an online and robust MOT approach, named Basketball-SORT, which focuses on the CMOO problems in basketball videos. To overcome the CMOO problem, instead of using the intersection-over-union-based (IoU-based) approach, we use the trajectories of neighboring frames based on the projected positions of the players. Our method designs the basketball game restriction (BGR) and reacquiring Long-Lost IDs (RLLI) based on the characteristics of basketball scenes, and we also solve the occlusion problem based on the player trajectories and appearance features. Experimental results show that our method achieves a Higher Order Tracking Accuracy (HOTA) score of 63.48$%$ on the basketball fixed video dataset and outperforms other recent popular approaches. Overall, our approach solved the CMOO problem more effectively than recent MOT algorithms.
Read more7/1/2024