Spatio-Temporal CNN baseline method for the Sports Video Task of MediaEval 2021 benchmark

TL;DR

This paper introduces a spatio-temporal CNN baseline for stroke detection and classification in sports videos, aiding participants in the MediaEval 2021 benchmark, and demonstrating competitive performance especially in detection.

Contribution

The paper presents a tailored spatio-temporal CNN baseline for sports video stroke detection and classification, serving as a reference for participants in the MediaEval 2021 benchmark.

Findings

Baseline outperforms other participants in detection task

Method tailored for specific subtasks

Highlights difficulty of stroke detection in sports videos

Abstract

This paper presents the baseline method proposed for the Sports Video task part of the MediaEval 2021 benchmark. This task proposes a stroke detection and a stroke classification subtasks. This baseline addresses both subtasks. The spatio-temporal CNN architecture and the training process of the model are tailored according to the addressed subtask. The method has the purpose of helping the participants to solve the task and is not meant to reach stateof-the-art performance. Still, for the detection task, the baseline is performing better than the other participants, which stresses the difficulty of such a task.