SportSense: User Interface for Sketch-Based Spatio-Temporal Team Sports Video Scene Retrieval

Authors
Lukas Probst, Ihab Al Kabary, Rufus Lobo, Fabian Rauschenbach, Heiko Schuldt, Philipp Seidenschwarz, Martin Rumo
Type
In Proceedings
Date
2018/3
Appears in
Proceedings of the IUI 2018 Workshop on User Interfaces for Spatial and Temporal Data Analysis (UISTDA '18)
Location
Tokyo, Japan
Abstract

In the last years, various sports have seen significant efforts in collecting large volumes of video, mainly for analytical purposes. These videos are tagged with events that include spatial and temporal information. In order to avoid that game analysts have to manually analyze large video collections, appropriate user interfaces are needed for finding characteristic video scenes. SportSense is a novel sketch-based video retrieval system tailored to the needs of sports video analysts which include spatio-temporal queries such as player movements, ball trajectories, or interactions between players. We present the user interface of SportSense that allows users to draw sketches of spatio-temporal events on the field and that visualizes the retrieved video scenes. We introduce the data model of SportSense, show the sketch-based spatio-temporal retrieval on the annotated videos, and present a qualitative evaluation exhibiting the effectiveness of the SportSense UI.