Pose Estimation and Mesh Generation

[UppuluriKalyani]

Project Title: Pose Estimation and Mesh Generation

Description:

This project aims to perform 3D human pose estimation from a 2D image input. The system will extract key human features, including heatmaps and silhouettes, to generate a 3D mesh of the human body. It will then render the mesh alongside projected silhouettes and keypoints, giving a detailed view of the pose. The project has applications in animation, AR/VR, and sports analytics.

Key Features:

2D human pose detection.

Generation of 3D human mesh models.

Use of PosePrior for pose estimation and ShapePrior for shape generation.

Rendering of the final silhouette and keypoints for visualization.

Tasks:

• Build a model to extract heatmaps and silhouettes from 2D images.

• Integrate the PosePrior and ShapePrior models for mesh generation.

• Render the 3D mesh and project silhouettes and keypoints.

• Optimize for performance and accuracy.

Technology Stack:

Deep Learning (CNNs)

Python

PyTorch / TensorFlow

OpenCV for rendering

Expected Output:

A system capable of taking a 2D image of a person and outputting a 3D mesh model of the human body with accurate pose estimation and projected keypoints.

[github-actions]

Thank you for creating this issue! 🎉 We'll look into it as soon as possible. In the meantime, please make sure to provide all the necessary details and context. Your contributions are highly appreciated! 😊

[shrehs]

Hello, hope your doing well. I would love to work on this issue. I'm part of GSSOC-24 Extended. If this could be assigned to me, that would be great. Thank you.

[UppuluriKalyani]

@shrehs go for it!

[github-actions]

Hello @UppuluriKalyani! Your issue #31 has been closed. Thank you for your contribution!