Enhancing Human Pose Estimation with Darknet-53 and Bidirectional LSTM Networks

Main Article Content

Surbhit Shukla, C. S. Raghuvanshi, Hari Om Sharan

Abstract

The accurate estimation of human poses is a fundamental problem in computer vision, with applications ranging from action recognition to human-computer interaction. This research explores the enhancement of human pose estimation accuracy through the integration of Darknet-53, a deep convolutional neural network, and Bidirectional Long Short-Term Memory (BiLSTM) networks. The proposed approach is evaluated on benchmark datasets, demonstrating significant improvements in human pose estimation accuracy. This paper provides a comprehensive analysis of the methodology and experimental results, highlighting the effectiveness of the Darknet-53 and BiLSTM combination.

Article Details

Section
Articles