Canada Spends Logo

Explorez notre analyse du budget 2026

Grant: Engage Grants Program

Concordia University

$25,000.00

Ice Hockey Shot Motion Extraction from Stereo Videos

EF 2016-17

Résumé
Reebok-CCM (legal name Sport Maska Inc.) would like to develop a computerized installation for sports shopsto capture and analyze a player's (customer) ice hockey shot actions. The larger goal is to automatically makesuitable hockey stick recommendations based on the player's shooting style. The scenario visualized is one inwhich the customer is asked to perform a few shots in a special room/area of the shop, which are captured.Since the goal is to put up such installations in various sports shops, low cost is essential. A high-end motioncapture installation with a multitude of high speed precision cameras is not therefore a viable solution. Anotherprincipal necessity is the ease with which a player's action can be captured and subsequently analyzed. For thisshort duration of this Engage project, we will be concerned primarily with capture and motion extraction.Motion extraction should yield player skeleton motion, along with stick and puck motion, all in 3D for later usein a recommendation engine. The big challenges in both capture and extraction are due to speed of the playeraction, and the very high speeds of stick and puck motion. Following a detailed discussion with technicalpersonnel from Reebok-CCM, the solution being proposed, for the capture problem is a calibrated stereocamera installation with appropriate lighting. The capture process is unencumbered, as markers or otherdevices are not required to be attached to the customer. Sticks and pucks will be pre-marked. For the motionextraction, we are proposing an innovative technique of tracking stereo pixel pairs in the captured video.Optical flow applied to the whole image is the current state of the art for tracking features in video. The mainproblem with optical flow, particularly with high speed motion is drift, making it difficult if not impossible totrack over the entire duration of the shot. In our proposed method, we will track individual points using a rigidlocal geometry prior. The geometry prior will not only enable entire shot duration tracking but will also enableus track even with small occlusions. The main deliverable of this project is a prototype installation for captureand motion extraction.
Programme
Engage Grants Program
Bénéficiaire
Emplacement
Quebec, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data

Tel que publié

Application id
501069-2016
Name nom
Mudur, SudhirPandurang
Department département
Computer Science and Software Engineering
Organization id
70
Institution établissement
Concordia University
Province en
Québec
Province fr
Québec
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2016
Competition year année de concours
2016
Award amount
25000
Program id
EGP
Program name en
Engage Grants Program
Program name fr
Programme de subventions d'engagement partenariat
Group en
Research partnerships programs
Group fr
Programmes de partenariat
Committee code
1552
Committee name en
Quebec Internal Decision Committee
Committee name fr
Comité de décision interne pour le Québec
Area of application code
1207
Area of application group en
Advancement of knowledge
Area of application group fr
Avancement des connaissances
Area of application en
Information, computer and communication technologies
Area of application fr
Technologie de l'information, des ordinateurs et des réseaux de communications
Research subject code
2700
Research subject group en
Information technology
Research subject group fr
Technologie de l'information
Research subject en
Information technology
Research subject fr
Technologie de l'information
Partie
1
Nb partie
1
Application title
Ice Hockey Shot Motion Extraction from Stereo Videos
Keyword
NSERC
Application summary
Reebok-CCM (legal name Sport Maska Inc.) would like to develop a computerized installation for sports shopsto capture and analyze a player's (customer) ice hockey shot actions. The larger goal is to automatically makesuitable hockey stick recommendations based on the player's shooting style. The scenario visualized is one inwhich the customer is asked to perform a few shots in a special room/area of the shop, which are captured.Since the goal is to put up such installations in various sports shops, low cost is essential. A high-end motioncapture installation with a multitude of high speed precision cameras is not therefore a viable solution. Anotherprincipal necessity is the ease with which a player's action can be captured and subsequently analyzed. For thisshort duration of this Engage project, we will be concerned primarily with capture and motion extraction.Motion extraction should yield player skeleton motion, along with stick and puck motion, all in 3D for later usein a recommendation engine. The big challenges in both capture and extraction are due to speed of the playeraction, and the very high speeds of stick and puck motion. Following a detailed discussion with technicalpersonnel from Reebok-CCM, the solution being proposed, for the capture problem is a calibrated stereocamera installation with appropriate lighting. The capture process is unencumbered, as markers or otherdevices are not required to be attached to the customer. Sticks and pucks will be pre-marked. For the motionextraction, we are proposing an innovative technique of tracking stereo pixel pairs in the captured video.Optical flow applied to the whole image is the current state of the art for tracking features in video. The mainproblem with optical flow, particularly with high speed motion is drift, making it difficult if not impossible totrack over the entire duration of the shot. In our proposed method, we will track individual points using a rigidlocal geometry prior. The geometry prior will not only enable entire shot duration tracking but will also enableus track even with small occlusions. The main deliverable of this project is a prototype installation for captureand motion extraction.