irolaina / FCRN-DepthPrediction

Deeper Depth Prediction with Fully Convolutional Residual Networks (FCRN)
BSD 2-Clause "Simplified" License
1.12k stars 312 forks source link

Running predict.py on multiple images #96

Open Saoussenl opened 3 years ago

Saoussenl commented 3 years ago

I get an error everytime I try to run the prediction on several images through a jupyter notebook.

Do you know how I can fix this?

I think it's better if you define a function load_model in predict.py to create model and load weights from the checkpoint and returns the model that could be passed later to the run function.

I'm trying to implement that but the tf.Session() seems to make it a bit complicated.