Optimal Choice of Motion Estimation Methods for Fine-Grained Action Classification with 3D Convolutional Networks - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2019

Optimal Choice of Motion Estimation Methods for Fine-Grained Action Classification with 3D Convolutional Networks

Résumé

Detecting and classifying human actions in videos is oneof the current challenges in visual content analysis and mining. This paper presents a method for performing a fine-grained classification of sport actions using a Siamese Spatio-Temporal Convolutional Neural Network (SSTCNN) model.This model takes RGB images and Optical Flow field as in-put data. Our first contribution is the comparison of different Optical flow methods and a study of their influence on the classification score. We also present different normalization methods for the optical flow that drastically impact results,boosting performances from 44% to 74% of accuracy. Our second contribution is the detection and classification of actions in videos performed using a sliding temporal window.It leads to a satisfying score of 81.3% over the whole dataset TTStroke-21.
Fichier principal
Vignette du fichier
ICIP2019_submission.pdf (822.63 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02326240 , version 1 (16-06-2020)

Identifiants

Citer

Pierre-Etienne Martin, Jenny Benois-Pineau, Renaud Peteri, Julien Morlier. Optimal Choice of Motion Estimation Methods for Fine-Grained Action Classification with 3D Convolutional Networks. 2019 IEEE International Conference on Image Processing (ICIP), Sep 2019, Taipei, Taiwan. pp.554-558, ⟨10.1109/ICIP.2019.8803780⟩. ⟨hal-02326240⟩
210 Consultations
69 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More