Load a trained model and generate novel view

[aeskandari68]

After completing the training process, I am wondering how to load the model and generate a view or video from it. Unfortunately, I couldn't find any script that handles inference or generates an image from the desired view in the repository. I came across a code snippet in the repository, but it doesn't seem to be implemented yet, and it's not exactly what I am looking for.

neuralangelo/projects/nerf/trainers/base.py

Line 158 in 94390b6

def dump_test_results(self, data_all, output_dir):

I found another code snippet (

neuralangelo/projects/nerf/trainers/nerf.py

Line 84 in 94390b6

def dump_test_results(self, data_all, output_dir):

)) and I am wondering if I can use it instead of the "base.py" one.

There is also another function for inference here:

neuralangelo/projects/neuralangelo/model.py

Line 74 in 94390b6

def inference(self, data):

I am feeling quite confused about how to generate and save an image from the desired viewpoint. Do you have any suggestions that could help me with this?