Recently my employer has tasked me with using image recognition with uploading photos instead of taking pictures with a camera.
Most tutorials online involve using a camera to capture images to train the model, so I was wondering if there was any support for uploading images.
Any help/advice is welcome. I am a beginner programmer, so there is little chance I could hard code this myself, so please keep that in mind.
Hi @bglaenger, please see this post: https://forums.developer.nvidia.com/t/help-with-image-recognition/127888/2
You should be able to do this without coding by putting your images in the expected directory structure of classification datasets. By “uploading”, do you mean that you have the files on a USB stick for example, and could simply copy/paste them into the dataset directories? Otherwise perhaps you could upload them with FTP or SCP.