What's difference of NN and curve? about deep_bilateral_network

[EricElmoznino]

Difference is in how the guide map is specified. Don’t recall the theoretical differences (they might mention them in the paper), but in my case “curve” always worked better. It’s a bit more annoying to implement in CoreML, but definitely doable.

Congrats on porting! Most annoying part of the process. If you don’t mind my asking, what’s the use-case and task you’re trying to learn?

[ewen1024]

In my case, the videos shoot on our machinces suffer from poor image quality(due to the limitation of lenses and sensors), so we'd use the model to 'brighten' the image to some extent.

By the way, any tests on the number of guide_pts?

[EricElmoznino]

No that I have access to, sorry. All tests I did are from around a year ago at a company I no longer work at.