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

Editing Interface for editing frames #22

Open
joeyklee opened this issue Jun 7, 2020 · 3 comments
Open

Editing Interface for editing frames #22

joeyklee opened this issue Jun 7, 2020 · 3 comments

Comments

@joeyklee
Copy link
Collaborator

joeyklee commented Jun 7, 2020

As a first pass for a simple editing interface for image frames. Sorry for the eye-sore of a wireframe, but working quickly to jot down some basic thoughts!

MANUAL-EDITING

@shiffman
Copy link
Owner

shiffman commented Jun 7, 2020

Something that could be useful (especially if a particular model drops a frame or two) is an option to "copy overlay from previous (or subsequent?) frame. I suppose instead of actually drawing on the video like a "stamp", we could consider all of the overlay elements to be manipulable elements like you can grow / shrink of adjust position for any given frame.

@joeyklee
Copy link
Collaborator Author

joeyklee commented Jun 7, 2020

@shiffman - Our approach at the moment is definitely a much simpler approach since we're just writing directly to the images. We could instead create a data model for storing the coordinates that would be used to generate canvas elements that could be manipulated. This would be mellow for the face bounding boxes, but for the bodyPix segmentation results ,this might be much harder. Or is there a good way to do this?

@shiffman
Copy link
Owner

shiffman commented Jun 8, 2020

Yes, I agree. I think this is probably a much later step and better served by learning what things people actually need to do with the videos. I agree that this would work much more easily for faces which can just be x,y,w,h objects.

Another simpler idea would be to save every frame as two images/canvases, the video and a transparent overly. This way a transparent overlay could be duplicated from one frame to another to save manual repetitious editing (for example if a face that stays in one place isn't being caught for many frames.)

@shiffman shiffman mentioned this issue Jun 8, 2020
7 tasks
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

No branches or pull requests

2 participants