Two-Streams: Dark and Light Networks with Graph Convolution for Action Recognition from Dark Videos (Student Abstract)

Authors

  • Saurabh Suman Indian Institute of Technology Jammu
  • Nilay Naharas Indian Institute of Technology Jammu
  • Badri Narayan Subudhi Indian Institute of Technology Jammu
  • Vinit Jakhetiya Indian Institute of Technology Jammu

DOI:

https://doi.org/10.1609/aaai.v37i13.27030

Keywords:

Computer Vision, Video Action Recognition, Dark Environment, Self-Calibrated Illumination, Two-Stream Network, Graph Convolutional Network, Convolutional Neural Network, Image Enhancement

Abstract

In this article, we propose a two-stream action recognition technique for recognizing human actions from dark videos. The proposed action recognition network consists of an image enhancement network with Self-Calibrated Illumination (SCI) module, followed by a two-stream action recognition network. We have used R(2+1)D as a feature extractor for both streams with shared weights. Graph Convolutional Network (GCN), a temporal graph encoder is utilized to enhance the obtained features which are then further fed to a classification head to recognize the actions in a video. The experimental results are presented on the recent benchmark ``ARID" dark-video database.
AAAI-23 Proceedings Cover

Downloads

Published

2024-07-15

How to Cite

Suman, S., Naharas, N., Subudhi, B. N., & Jakhetiya, V. (2024). Two-Streams: Dark and Light Networks with Graph Convolution for Action Recognition from Dark Videos (Student Abstract). Proceedings of the AAAI Conference on Artificial Intelligence, 37(13), 16340-16341. https://doi.org/10.1609/aaai.v37i13.27030

Issue

Vol. 37 No. 13: AAAI-23 Special Programs, IAAI-23, EAAI-23, Student Papers and Demonstrations

Section

AAAI Student Abstract and Poster Program