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
ML-based Autotracker #33
Comments
Hi again Kevin, you sound like you really know your stuff and have some
great ideas. Is this something you've already worked with in another
context? Again, I know nothing about current ML methods of object tracking
but certainly a better autotracker is always desirable. Tell us more!
…On Sat, Nov 14, 2020 at 10:04 AM Kevin Tan ***@***.***> wrote:
Hi there again! This is a RFC to see if people would be interested in an
ML-based auto-tracking tool. It seems like the current auto-tracker uses a
slightly brittle template-matching system and I've been having trouble with
it especially when the lighting in the image in variable and an object
(e.g. a bouncy ball) enters and exits a shadowy region. I think a lot of
modern object detection/localization ML methods might be able to
dramatically increase the fidelity of the auto-tracking system. Let me know
what you think!
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
<#33>, or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AAXIBFH6ZBEZJ73C7TJADEDSP3BC5ANCNFSM4TVVXEAA>
.
|
Hi Doug, again, sorry for the untimely response. Happy Thanksgiving! I haven't worked on specifically object tracking but I have worked on some machine learning projects in the past. I'm more of a computer systems guy than an artificial intelligence person, but I can ask around to see what I might be able to find out. I'll touch base when I hear back! |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Hi there again! This is a RFC to see if people would be interested in an ML-based auto-tracking tool. It seems like the current auto-tracker uses a slightly brittle template-matching system and I've been having trouble with it especially when the lighting in the image in variable and an object (e.g. a bouncy ball) enters and exits a shadowy region. I think a lot of modern object detection/localization ML methods might be able to dramatically increase the fidelity of the auto-tracking system. Let me know what you think!
The text was updated successfully, but these errors were encountered: