Open DonaldTsang opened 6 years ago
Better to use DNN, it gives better accuracy. If you need to use classic CV descriptor first you need to know how they work, cause they very specific and work not in all cases. But yes u can use Kaze.
Kindly yours, Andrey Nikishaev
Areas ML/DS/CV/Soft Dev/BizDev/Growth Hacking/Customer Rel/IT LinkedIn http://ua.linkedin.com/in/creotiv GitHub http://github.com/creotiv Slideshare https://www.slideshare.net/anikishaev/ Skype creotiv.in.ua Mobile +380632410666
On Thu, Oct 25, 2018 at 12:52 PM Donald Tsang notifications@github.com wrote:
Can KAZE/AKAZE be used in in image databases for finding possible duplicates? Because I would like to add that to https://github.com/hydrusnetwork/hydrus in order to enhance the image deduplication and sourcing capabilities
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/creotiv/computer_vision/issues/2, or mute the thread https://github.com/notifications/unsubscribe-auth/AAPRDUsY0xLxuvLJlRpXA6sXK553fdEcks5uoYn1gaJpZM4X56Ev .
@creotiv My purpose is to use a type of image feature "hash" to find duplicate and partially duplicate images with affine transformations and cropping (and possibly combined images similar to mosaics)
Can KAZE/AKAZE be used in in image databases for finding possible duplicates? Because I would like to add that to https://github.com/hydrusnetwork/hydrus in order to enhance the image deduplication and sourcing capabilities