Tanuj Jain
Tanuj Jain
Facing the same issue on an ubuntu GPU machine. Any resolution yet?
@thorade You can check out the python package [imagededup](https://github.com/idealo/imagededup) that has the capability to find duplicates using perceptual hash.
Currently, imagededup is not well suited for the scale of images you have. We have tried upto 60K images on a colab notebook similar in configuration to yours and it...
@datitran We would no longer be able to support py3.6 after this fix is shipped. Just want to confirm if you're onboard with that?
Hi, we released a new version that gets rid of tensorflow and replaces it with pytorch. Please feel free to test it out.
Hi, new release of imagededup is out that doesn't depend upon tensorflow anymore. That should address installation issues.
CNNs exhibit a good degree of rotation invariance. So, you should try CNN method instead of hashes for obtaining duplicates in this case. I also encourage you to have a...
Hi We're trying to resolve some dependency issues before floating the new version. Meanwhile, you could install it from source using the dev branch.
Hi, new version of imagededup should work fine. Please try it out.
Apparently, Pillow==6.2.2 addresses the mentioned security issues: https://pillow.readthedocs.io/en/stable/releasenotes/6.2.2.html More generally, AHash generates different hashes with the latest Pillow release (8.0.0), which isn't ideal. Would need to investigate exactly why that...