Baseline Method for the Sport Task of MediaEval 2022 with 3D CNNs using Attention Mechanisms

Pierre-Etienne Martin
2/6/2023
cs.CVcs.AIcs.LG

No official code found. Best implementation: ccp-eva/SportTaskME22 (6 stars, last commit 2024-05-14).

Abstract

This paper presents the baseline method proposed for the Sports Video task part of the MediaEval 2022 benchmark. This task proposes two subtasks: stroke classification from trimmed videos, and stroke detection from untrimmed videos. This baseline addresses both subtasks. We propose two types of 3D-CNN architectures to solve the two subtasks. Both 3D-CNNs use Spatio-temporal convolutions and attention mechanisms. The architectures and the training process are tailored to solve the addressed subtask. This baseline method is shared publicly online to help the participants in their investigation and alleviate eventually some aspects of the task such as video processing, training method, evaluation and submission routine. The baseline method reaches 86.4% of accuracy with our v2 model for the classification subtask. For the detection subtask, the baseline reaches a mAP of 0.131 and IoU of 0.515 with our v1 model.

View on arXivView PDF

Code Implementations(1)

ccp-eva/SportTaskME22Community100%
61PythonApr 28, 20222 years ago

Similar Papers

Effects of Different Attention Mechanisms Applied on 3D Models in Video Classification

83% match2026

Fine-Grained Action Detection with RGB and Pose Information using Two Stream Convolutional Networks

82% match2023

Sport Task: Fine Grained Action Detection and Classification of Table Tennis Strokes from Videos for MediaEval 2022

81% match1 repo62023

CNNtention: Can CNNs do better with Attention?

80% match1 repo62024

Unifying Global and Local Scene Entities Modelling for Precise Action Spotting

80% match1 repo102024

Video4MRI: An Empirical Study on Brain Magnetic Resonance Image Analytics with CNN-based Video Classification Frameworks

80% match2023

Cite this paper

@article{martin2023baseline,
  title  = {Baseline Method for the Sport Task of MediaEval 2022 with 3D CNNs using Attention Mechanisms},
  author = {Pierre-Etienne Martin},
  year   = {2023},
  eprint = {2302.02752},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2302.02752v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2302.02752.svg)](https://codeofpaper.com/paper/2302.02752)

Discussion