Skip to content

For Wolfram Summer School 2023, I am working estimating 3D pose in form of human skeletons and 3D meshes from video and audio data using neural networks.

Notifications You must be signed in to change notification settings

Fizza-Rubab/WSS23-3D-Pose-Estimation

Repository files navigation

WSS23 - 3D Human Pose Estimation

3D Human Pose Estimation using Wolfram Language

For Wolfram Summer School 2023, I worked on estimating 3D pose in form of human skeletons and 3D meshes from video and images using neural networks.

Objective

This project aims to leverage advanced computer vision techniques to accurately estimate the 3D positions of 17 keypoints on the human body and create visually appealing visualizations of the estimated pose in a 3-dimensional display.

Approach

This project implemented a two-step approach for 3D human pose estimation. The CenterNet model from the Wolfram Neural Network repository was employed to estimate the 2D coordinates of 17 keypoints on the human body. The MiDaS monocular depth estimation model was utilized by importing it using ONNX to obtain depth information. By combining the 2D coordinates with the estimated depths, the body joint linkages were established and the pose was visualized in 3D. We tested the pipeline on various single person images and videos, and it demonstrated comparable results and visualizations to state-of-the-art technologies.

3D Human Pose Estimation using Wolfram Language

Future Directions:

  1. Multi-person support: Extend the project to estimate 3D keypoints for multiple individuals in an image while maintaining their separation as distinct human skeletons.
  2. Occlusion handling: Explore the development of an occlusion-resistant model capable of estimating keypoints even when body parts are partially hidden.
  3. Video optimization: Investigate ways to enhance and implement the 3D estimation pipeline for video data.
  4. Mesh Regression: Consider further work on converting 3D poses into a standard human mesh using the obtained 3D points.

Gallery

Here are some interesting results on images. 3D Human Pose Estimation using Wolfram Language 3D Human Pose Estimation using Wolfram Language 3D Human Pose Estimation using Wolfram Language

More details can be found in this Wolfram Community Post

About

For Wolfram Summer School 2023, I am working estimating 3D pose in form of human skeletons and 3D meshes from video and audio data using neural networks.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published