Issue max res 720 * 1280 ? about boostingmonoculardepth

[StevenLi-phoenix]

device: cuda
Namespace(Final=True, R0=False, R20=False, colorize_results=False, data_dir='/content/inputs', depthNet=2, max_res=inf, net_receptive_field_size=None, output_dir='/content/drive/MyDrive/outputs_leres/', output_resolution=1, pix2pixsize=1024, savepatchs=0, savewholeest=0)
----------------- Options ---------------
Final: True [default: False]
R0: False
R20: False
aspect_ratio: 1.0
batch_size: 1
checkpoints_dir: ./pix2pix/checkpoints
colorize_results: False
crop_size: 672
data_dir: /content/inputs [default: None]
dataroot: None
dataset_mode: depthmerge
depthNet: 2 [default: None]
direction: AtoB
display_winsize: 256
epoch: latest
eval: False
generatevideo: None
gpu_ids: 0
init_gain: 0.02
init_type: normal
input_nc: 2
isTrain: False [default: None]
load_iter: 0 [default: 0]
load_size: 672
max_dataset_size: 10000
max_res: inf
model: pix2pix4depth
n_layers_D: 3
name: void
ndf: 64
netD: basic
netG: unet_1024
net_receptive_field_size: None
ngf: 64
no_dropout: False
no_flip: False
norm: none
num_test: 50
num_threads: 4
output_dir: /content/drive/MyDrive/outputs_leres/ [default: None]
output_nc: 1
output_resolution: None
phase: test
pix2pixsize: None
preprocess: resize_and_crop
savecrops: None
savewholeest: None
serial_batches: False
suffix:
verbose: False
----------------- End -------------------
initialize network with normal
loading the model from ./pix2pix/checkpoints/mergemodel/latest_net_G.pth
start processing
processing image 0 : 0
wholeImage being processed in : 2688
Adjust factor is: 1.0
Selecting patchs ...
Target resolution: (3024, 5376, 3)
Dynamicly change merged-in resolution; scale: 0.23809523809523808
Resulted depthmap res will be : (720, 1280)
patchs to process: 55
processing patch 0 | [ 0 0 693 693]
processing patch 1 | [ 80 0 693 693]
processing patch 2 | [160 0 693 693]

[StevenLi-phoenix]

I mean actually using the model to predict the result. Not just using resize.
Any suggestions?

[miangoleh]

Can you share the command you use to run the model?

[StevenLi-phoenix]

!python run.py --Final --data_dir /content/inputs --output_dir /content/outputs_leres/ --depthNet 2

May I need to add specific parameter that I need to add?

[StevenLi-phoenix]

!python run.py --Final --data_dir /content/inputs --output_dir /content/outputs_leres/ --depthNet 2

May I need to add specific parameter that I need to add?

Just running the code in the co-lab example

[miangoleh]

It seems that your input image resolution is (720, 1280). For consistency and avoiding too large output image files we are resizing our methods output to input original resolution [ in this case from (3024, 5376, 3) to (720, 1280, 3) ]. If you dont want the output to be resized to input resolution consider using --output_resolution 0.

[StevenLi-phoenix]

ok.
I think understand what's happening.

The model is processing the image in it input resolution and to avoid too large output, it automatically resize the image after processing.

Before:

After adding the params:

That's working.

[StevenLi-phoenix]

Maybe there is a max resolution limit? I definitely uploaded a super high resolution image
This one: (from https://kids.nationalgeographic.com/geography/states/article/new-york)

size: 3072 × 1728

Thank you for your help. I appreciate it very much.