Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Pose #56

Closed
wants to merge 33 commits into from
Closed

Pose #56

wants to merge 33 commits into from

Conversation

john-rocky
Copy link
Contributor

@john-rocky john-rocky commented Aug 12, 2024

@ambitious-octopus

I've added pose functionality to the app and committed to the pose branch.
This branch is based on the branch that
added human and segmentation.
As always, please let me know if there are any issues that need to be resolved during the integration process (e.g. maintaining model references).

Thanks in advance.

IMG_3582

🛠️ PR Summary

Made with ❤️ by Ultralytics Actions

🌟 Summary

Enhanced the YOLO iOS app to support pose estimation, added new visualization capabilities, and improved UI modularity.

📊 Key Changes

  • Pose Estimation Support:
    • Integrated YOLOv8 pose models (n, s, m, l, x) for human pose estimation.
    • Added PostProcessPose.swift for processing pose inference results and rendering skeletons.
    • Updated existing model-switching logic to include detection and pose model modes.
  • UI Enhancements:
    • Introduced an overlay layer to display pose keypoints and skeletons.
    • Updated interface to switch between "Detection" and "Pose" modes via a segmented control.
    • Modified bounding box color styling using Ultralytics’ color palette for consistent visuals.
  • Utilities:
    • Added Colors.swift with color definitions, including pose palettes for customization.
    • Improved non-max suppression and scaling logic for better bounding box accuracy.
  • Versioning:
    • Updated app version to 8.3.0 and project build version to 2.

🎯 Purpose & Impact

  • Enhanced Functionality: Pose estimation expands the app’s use cases beyond object detection, enabling human activity recognition and gesture analysis.
  • Improved Visualization: Users can now visually interpret pose estimations through detailed skeleton overlays, providing clearer insights.
  • Better UI Experience: Seamless switching between detection and pose modes improves adaptability for diverse scenarios.
  • Consistency: Alignment of visual elements with Ultralytics branding ensures a cohesive user experience.

🎉 Users can now detect and analyze human poses directly on their iOS devices with enhanced visualizations!

@glenn-jocher glenn-jocher added APP dependencies Dependencies and packages enhancement New feature or request labels Aug 12, 2024
Copy link
Member

@ambitious-octopus ambitious-octopus left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hey @john-rocky , could you please review this PR and remove any unnecessary elements? I’d like to ensure we’re focusing solely on items related to pose estimation. Thanks!

@john-rocky
Copy link
Contributor Author

john-rocky commented Aug 21, 2024

@ambitious-octopus
Understood. I will modify this branch to a simple code that adds pose function to the main branch. Please wait a moment.

@john-rocky
Copy link
Contributor Author

@ambitious-octopus
I removed unnecessary elements. Currently, this branch only adds the pause function to the main branch.

Copy link
Member

@ambitious-octopus ambitious-octopus left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hey @john-rocky it seems that there are still some merge conflicts. Can you please resolve them?

@john-rocky
Copy link
Contributor Author

@ambitious-octopus
I fixed the conflicts.
I tested this branch. This works fine.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
dependencies Dependencies and packages enhancement New feature or request
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants