kacperkan / conerf Goto Github PK

Resolved, but there are a ton of things to manage with versions.

(conerf) : /Dev/conerf$ echo $DATASET_PATH
/home/user/Dev/conerf/datasets/captures-camera-ready/face-2-attributes/
(conerf): /Dev/conerf$ echo $EXPERIMENT_PATH
/home/user/Dev/conerf/res/
(conerf) : /Dev/conerf$ python train.py --base_folder $EXPERIMENT_PATH --gin_bindings="data_dir='$DATASET_PATH'" --gin_configs configs/test_local_attributes.gin
I0910 13:22:44.964829 139876748982080 train.py:152] *** Starting experiment
I0910 13:22:44.964997 139876748982080 train.py:156] *** Loading Gin configs from: ['configs/test_local_attributes.g']
I0910 13:22:44.965285 139876748982080 resource_reader.py:50] system_path_file_exists:configs/test_local_attributes.g
Traceback (most recent call last):
File "train.py", line 461, in
app.run(main)
File "/home/user/anaconda3/envs/conerf/lib/python3.8/site-packages/absl/app.py", line 312, in run
_run_main(main, args)
File "/home/user/anaconda3/envs/conerf/lib/python3.8/site-packages/absl/app.py", line 258, in _run_main
sys.exit(main(argv))
File "train.py", line 157, in main
gin.parse_config_files_and_bindings(
File "/home/user/anaconda3/envs/conerf/lib/python3.8/site-packages/gin/config.py", line 2497, in parse_config_files_and_bindings
includes_and_imports = parse_config_file(config_file, skip_unknown)
File "/home/user/anaconda3/envs/conerf/lib/python3.8/site-packages/gin/config.py", line 2448, in parse_config_file
if existence_check(config_file_with_prefix):
File "/home/user/anaconda3/envs/conerf/lib/python3.8/site-packages/gin/resource_reader.py", line 52, in system_path_file_exists
path = _parse_config_path(config_path)
File "/home/user/anaconda3/envs/conerf/lib/python3.8/site-packages/gin/resource_reader.py", line 89, in _parse_config_path
path = os.path.join(os.path.dirname(file_sys_path), filename)
File "/home/user/anaconda3/envs/conerf/lib/python3.8/posixpath.py", line 152, in dirname
p = os.fspath(p)
TypeError: expected str, bytes or os.PathLike object, not NoneType

Thanks for your work!

I'm curious about how to generate unseen attribute combinations as mentioned in the paper.
Do you generate it by fixing one single latent code β and changing those attributes?

In my opinion, this shouldn't work: since the deformed coordinates (K(x) in the paper) depend only on β, if you have changed the canonical space by changing the value of those attributes but K(x) remains the same, the K(x) may not be feasible on the canonical area that has been changed.

So I think you may have used several latent codes β to generate unseen attribute combinations. But this requires the images corresponding to those β to be well aligned, right? So this brings some constraints to this work?

Thanks!

Thanks for your fantastic work! I have a question about Colmap.
I'm trying to run CoNerf on my own captured data (with a mobile phone).
But when I try to use colmap to predict my camera parameters, the results are bad.
So could you please tell me your colmap configuration? Or any advice on how to use a mobile phone to record data (any record setting?)?

Thanks!

I saw your updates in README are about hyperNerf but the title is conerf.

The link to the 52GB dataset doesn't work