search

bourdakos1 / Custom-Object-Detection

Custom Object Detection with TensorFlow
https://medium.freecodecamp.org/tracking-the-millenium-falcon-with-tensorflow-c8c86419225e
MIT License
347 stars 181 forks source link

My own images fails #29

Closed gamcoh closed 5 years ago

gamcoh commented 6 years ago

Hi nick,

I'm new to tensorflow so this may be a dumb question, but every output images that the program makes is resized to 558 of height and i think maybe that's why the program doesn't recognise features in my test images.

Maybe you can help?

bourdakos1 commented 5 years ago

Hi @gamcoh, this repo relies on a dated version of the TensorFlow api. We've moved to a more future proof version here: https://github.com/cloud-annotations/training I encourage you to try it out and reopen this issue there if you are still running into problems

bourdakos1 commented 5 years ago

This could definitely be obscuring features, just keep this in mind when gathering data. If the desired object is only a few pixels in a couple thousand pixel image they might disappear when shrunk. also when one dimension is much longer than the other features may get lost.

bourdakos1 commented 5 years ago

if you want to run the image at full resolution you can try breaking the image into pieces