Mutual modality learning for video action classification

The construction of models for video action classification progresses rapidly. However, the performance of those models can still be easily improved by ensembling with the same models trained on different modalities (e.g. Optical flow). Unfortunately, it is computationally expensive to use several m...

Full description

Saved in:
Bibliographic Details
Published inKompʹûternaâ optika Vol. 47; no. 4; pp. 637 - 649
Main Authors Komkov, S.A., Dzabraev, M.D., Petiushko, A.A.
Format Journal Article
LanguageEnglish
Published Samara National Research University 01.08.2023
Subjects
mutual learning
optical flow
video action classification
video labeling
video recognition
Online AccessGet full text

Cover

More Information
Summary:The construction of models for video action classification progresses rapidly. However, the performance of those models can still be easily improved by ensembling with the same models trained on different modalities (e.g. Optical flow). Unfortunately, it is computationally expensive to use several modalities during inference. Recent works examine the ways to integrate advantages of multi-modality into a single RGB-model. Yet, there is still room for improvement. In this paper, we explore various methods to embed the ensemble power into a single model. We show that proper initialization, as well as mutual modality learning, enhances single-modality models. As a result, we achieve state-of-the-art results in the Something-Something-v2 benchmark.
ISSN:0134-2452
2412-6179
DOI:10.18287/2412-6179-CO-1277