Semi-Coupled Two-Stream Fusion ConvNets for Action Recognition at Extremely Low Resolutions

TL;DR

This paper introduces a semi-coupled filter-sharing network that enhances action recognition in extremely low-resolution videos by leveraging high-resolution training data and fusing spatial-temporal information.

Contribution

It proposes a novel semi-coupled network architecture and fusion methods specifically designed for eLR videos, improving recognition accuracy over existing approaches.

Findings

Achieved 93.7% accuracy on IXMAS dataset

Achieved 29.2% accuracy on HMDB dataset

Outperformed state-of-the-art methods at eLR

Abstract

Deep convolutional neural networks (ConvNets) have been recently shown to attain state-of-the-art performance for action recognition on standard-resolution videos. However, less attention has been paid to recognition performance at extremely low resolutions (eLR) (e.g., 16 x 12 pixels). Reliable action recognition using eLR cameras would address privacy concerns in various application environments such as private homes, hospitals, nursing/rehabilitation facilities, etc. In this paper, we propose a semi-coupled filter-sharing network that leverages high resolution (HR) videos during training in order to assist an eLR ConvNet. We also study methods for fusing spatial and temporal ConvNets customized for eLR videos in order to take advantage of appearance and motion information. Our method outperforms state-of-the-art methods at extremely low resolutions on IXMAS (93.7%) and HMDB (29.2%) datasets.