Skip to main content

DroneAttention: Sparse weighted temporal attention for drone-camera based activity recognition.

Yadav, S. K., Luthra, A., Pahwa, E., Tiwari, K., Rathore, H., Pandey, H. and Corcoran, P., 2023. DroneAttention: Sparse weighted temporal attention for drone-camera based activity recognition. Neural Networks, 159, 57-69.

Full text available as:

[img]
Preview
PDF
Drone_One_Elsevier.pdf - Accepted Version
Available under License Creative Commons Attribution Non-commercial No Derivatives.

17MB

Official URL: https://www.sciencedirect.com/journal/neural-netwo...

DOI: 10.1016/j.neunet.2022.12.005

Abstract

Human activity recognition (HAR) using drone-mounted cameras has attracted considerable interest from the computer vision research community in recent years. A robust and efficient HAR system has a pivotal role in fields like video surveillance, crowd behavior analysis, sports analysis, and human-computer interaction. What makes it challenging are the complex poses, understanding different viewpoints, and the environmental scenarios where the action is taking place. To address such complexities, in this paper, we propose a novel Sparse Weighted Temporal Attention (SWTA) module to utilize sparsely sampled video frames for obtaining global weighted temporal attention. The proposed SWTA is comprised of two parts. First, temporal segment network that sparsely samples a given set of frames. Second, weighted temporal attention, which incorporates a fusion of attention maps derived from optical flow, with raw RGB images. This is followed by a basenet network, which comprises a convolutional neural network (CNN) module along with fully connected layers that provide us with activity recognition. The SWTA network can be used as a plug-in module to the existing deep CNN architectures, for optimizing them to learn temporal information by eliminating the need for a separate temporal stream. It has been evaluated on three publicly available benchmark datasets, namely Okutama, MOD20, and Drone-Action. The proposed model has received an accuracy of 72.76%, 92.56%, and 78.86% on the respective datasets thereby surpassing the previous state-of-the-art performances by a margin of 25.26%, 18.56%, and 2.94%, respectively.

Item Type:Article
ISSN:0893-6080
Uncontrolled Keywords:Action Recognition; Sparse Weighted Temporal Attention; Drone Vision
Group:Faculty of Science & Technology
ID Code:37900
Deposited By: Symplectic RT2
Deposited On:15 Dec 2022 16:38
Last Modified:13 Dec 2023 01:08

Downloads

Downloads per month over past year

More statistics for this item...
Repository Staff Only -