-
Notifications
You must be signed in to change notification settings - Fork 481
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
SIFT alternative #52
Comments
It is definitely doable to switch to another feature, although one it looks
like AKAZE uses binary descriptors, rather than integer-valued descriptors,
which means that for that particular feature would need to use another
approximate nearest neighbors library. In general, switching to another
feature would require changes to the feature I/O code and possibly other
places where for instance 128-dimensional descriptors are assumed.
Depending on the specific problem you are trying to solve, you could also
consider other SfM/SLAM packages and see if they have dependencies on
different features.
Noah
On Mon, May 22, 2017 at 7:43 PM Kevin Chiu ***@***.***> wrote:
SIFT is patented and makes using Bundler difficult for real world
purposes. Would it be straightforward to replace SIFT with, for example,
AKAZE from OpenCV?
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
<#52>, or mute the thread
<https://github.com/notifications/unsubscribe-auth/ABt6qwnMpngSF8BeYdWXs8E8hjhdZGAEks5r8h28gaJpZM4NjA10>
.
--
…-------------------------------------------------------------------
Noah Snavely
Associate Professor Email: [email protected]
Dept. of Computer Science Phone: (607) 255 4280 <(607)%20255-4280>
Cornell University URL : www.cs.cornell.edu/~snavely
307 Gates Hall
Ithaca, NY 14853
-------------------------------------------------------------------
|
For instance, I think Theia might support AKAZE:
https://github.com/sweeneychris/TheiaSfM
Noah
…On Wed, May 24, 2017 at 9:47 AM Noah Snavely ***@***.***> wrote:
It is definitely doable to switch to another feature, although one it
looks like AKAZE uses binary descriptors, rather than integer-valued
descriptors, which means that for that particular feature would need to use
another approximate nearest neighbors library. In general, switching to
another feature would require changes to the feature I/O code and possibly
other places where for instance 128-dimensional descriptors are assumed.
Depending on the specific problem you are trying to solve, you could also
consider other SfM/SLAM packages and see if they have dependencies on
different features.
Noah
On Mon, May 22, 2017 at 7:43 PM Kevin Chiu ***@***.***>
wrote:
> SIFT is patented and makes using Bundler difficult for real world
> purposes. Would it be straightforward to replace SIFT with, for example,
> AKAZE from OpenCV?
>
> —
> You are receiving this because you are subscribed to this thread.
> Reply to this email directly, view it on GitHub
> <#52>, or mute the thread
> <https://github.com/notifications/unsubscribe-auth/ABt6qwnMpngSF8BeYdWXs8E8hjhdZGAEks5r8h28gaJpZM4NjA10>
> .
>
--
-------------------------------------------------------------------
Noah Snavely
Associate Professor Email: ***@***.***
Dept. of Computer Science Phone: (607) 255 4280 <(607)%20255-4280>
Cornell University URL : www.cs.cornell.edu/~snavely
307 Gates Hall
Ithaca, NY 14853
-------------------------------------------------------------------
--
-------------------------------------------------------------------
Noah Snavely
Associate Professor Email: [email protected]
Dept. of Computer Science Phone: (607) 255 4280
Cornell University URL : www.cs.cornell.edu/~snavely
307 Gates Hall
Ithaca, NY 14853
-------------------------------------------------------------------
|
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
SIFT is patented and makes using Bundler difficult for real world purposes. Would it be straightforward to replace SIFT with, for example, AKAZE from OpenCV?
The text was updated successfully, but these errors were encountered: