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

Add docs for Handpose model #1067

Merged
merged 2 commits into from
Oct 15, 2020
Merged

Add docs for Handpose model #1067

merged 2 commits into from
Oct 15, 2020

Conversation

bomanimc
Copy link
Member

This PR introduces a documentation page for the Handpose model. While here, I also made some fixes to the Facemesh documentation to correct some typos I had accidentally left in that documentation.

Screen Shot 2020-09-30 at 2 44 45 PM (2)

Copy link
Member

@shiffman shiffman left a comment

Choose a reason for hiding this comment

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

This is so exciting! I am looking forward to trying to combine this with ml5.neuralNetwork() for training a hand classification model. Rock, paper, scissors example here we come!

@@ -51,7 +51,7 @@ const facemesh = ml5.facemesh(?video, ?options, ?callback);
detectionConfidence: 0.9, // Threshold for discarding a prediction. Defaults to 0.9.
maxFaces: 10, // The maximum number of faces detected in the input. Should be set to the minimum number for performance. Defaults to 10.
iouThreshold: 0.3, // A float representing the threshold for deciding whether boxes overlap too much in non-maximum suppression. Must be between [0, 1]. Defaults to 0.3.
scoreThreshold: 0.75, // defaults to 0.75
scoreThreshold: 0.75, // A threshold for deciding when to remove boxes based on score in non-maximum suppression. Defaults to 0.75
Copy link
Member

Choose a reason for hiding this comment

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

Noting here that score in non-maximum suppression is not familiar terminology. Looking it up myself I just watched a bit of this video https://www.coursera.org/lecture/convolutional-neural-networks/non-max-suppression-dvrjH. We could refer to a resource and/or simplify the language to something like:

// threshold for removing multiple (likely duplicate) detections

or

// threshold for removing multiple (likely duplicate) detections based on a "non-maximum suppression" algorithm. 

Copy link
Contributor

Choose a reason for hiding this comment

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

+1 for // threshold for removing multiple (likely duplicate) detections based on a "non-maximum suppression" algorithm.

Copy link
Member Author

Choose a reason for hiding this comment

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

Sounds good to me!

## Tutorials

No tutorials yet - contribute one today!

Copy link
Contributor

Choose a reason for hiding this comment

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

We should add in the placeholder for the Data and Model Provenance work that @ellennickles has done for the other features: e.g. https://github.com/ml5js/ml5-library/pull/1047/files#diff-8eef2538dbb4d1c92215a82d44c7fa2cR166

Copy link
Member Author

Choose a reason for hiding this comment

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

Added!

Copy link
Contributor

@joeyklee joeyklee left a comment

Choose a reason for hiding this comment

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

LGTM! I added one non-blocking comment regarding the data and model provenance work from @ellennickles and added a +1 to Dan's suggestion about a comment to explain the non-max suppression parameter.

Nice work!

… non-maximum suppression on the Facemesh & Handpose docs
@bomanimc bomanimc merged commit 273461c into development Oct 15, 2020
@bomanimc bomanimc deleted the bomani.handpose-docs branch October 15, 2020 02:07
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

3 participants