Bipose: Human Pose Estimation using ResNet-50 with BiLSTMs

Abstract

Human pose estimation is a critical task in computer vision that involves detecting and locating the positions of multiple body joints in images or videos. In this research paper, we propose a novel approach for human pose estimation using a combination of ResNet-50, a popular deep convolutional neural network, and bidirectional Long Short-Term Memory (BiLSTM) units. The proposed model aims to capture both local and temporal dependencies, enabling accurate and robust pose estimation even in complex and dynamic scenarios. We conduct extensive experiments on benchmark datasets to evaluate the effectiveness of our approach and compare it with state-of-the-art methods. The results demonstrate the superiority of our model in accurately estimating human poses.