lwt3437 / lanet Goto Github PK

Hello,

MSU Graphics & Media Lab Video Group has recently launched HDR Video Reconstruction Benchmark.

If you are interested in participating, you can add your algorithm following the submission steps: Submit for MSU HDR Video Reconstruction Benchmark

We would be grateful for your feedback on our work!

Hi, thank you for this amazing work!
And I have some questions about the LDR and HDR pre-processing to ensure scale-invariant.

In

Besides, in the loading procedure (

Hope for your reply and thank you again.

hello, I have one question about the dataset. I opened the source you supply in the paper, such as https://polyhaven.com/hdris. But the website only has panoramic HDR image, and there is no LDR image. May I ask you how to get LDR panoramic image from HDR？Thanks a lot.

Hello! I am trying to revitalise this network as it seems very good at producing image based lighting from panoramas.

However even after building the environment correctly in a runpod environment using a GTX3090 and 126GB of RAM it all runs extremely slowly with very minimal utilisation of any GPU memory. Currently the test image provided in ./test/pano1.png always hangs on "processing" instead of being processed in a few seconds.

GPU utilisation also seems to be very low, at around ~5% even though I am using tensorflow-gpu==1.13.1 and I can confirm that CUDA is available

Do you have any tips for getting it to run inference fast or if there are any similar more up to date projects?

WARNING:tensorflow:From j:\TIM\HDR\ITM_method\LANET\LANet-main\LANet-main\LANet\src\model.py:161: start_queue_runners (from tensorflow.python.training.queue_runner_impl) is deprecated and will be removed in a future version.

Because this function was deprecated it caused the program to hang and not run. Is there any way to fix this? If you can answer this, thank you for your answer.

Hi
Wonderful work. While running the code for training im getting this error:

Traceback (most recent call last):
  File "/home/aakash.kt/anaconda3/envs/tf-gpu/lib/python3.6/threading.py", line 916, in _bootstrap_inner
    self.run()
  File "/home/aakash.kt/anaconda3/envs/tf-gpu/lib/python3.6/threading.py", line 864, in run
    self._target(*self._args, **self._kwargs)
  File "/home/aakash.kt/Pano_HDR/LANet/LANet/src/model.py", line 334, in enqueue_frames
    coord.request_stop(e)
  File "/home/aakash.kt/anaconda3/envs/tf-gpu/lib/python3.6/site-packages/tensorflow/python/training/coordinator.py", line 213, in request_stop
    six.reraise(*sys.exc_info())
  File "/home/aakash.kt/anaconda3/envs/tf-gpu/lib/python3.6/site-packages/six.py", line 719, in reraise
    raise value
  File "/home/aakash.kt/Pano_HDR/LANet/LANet/src/model.py", line 320, in enqueue_frames
    self.sess.run(self.q_frames.close())
  File "/home/aakash.kt/anaconda3/envs/tf-gpu/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 877, in run
    run_metadata_ptr)
  File "/home/aakash.kt/anaconda3/envs/tf-gpu/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1100, in _run
    feed_dict_tensor, options, run_metadata)
  File "/home/aakash.kt/anaconda3/envs/tf-gpu/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1272, in _do_run
    run_metadata)
  File "/home/aakash.kt/anaconda3/envs/tf-gpu/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1291, in _do_call
    raise type(e)(node_def, op, message)
tensorflow.python.framework.errors_impl.CancelledError: Session has been closed.

I am unable to figure this out, please help @LWT3437

Hi,May I ask you a question? in your article，‘HDR images stored in relative luminance are scale-invariant, which
means the HDR images will hold the same information when multiplied by any positive real number. Based on this observation,
we propose a novel normalization method called " HDR calibration " for HDR images stored in relative luminance, calibrating
HDR images into a similar luminance scale according to the LDR images’ ,but in wiki,the definition of HDR is 'Information stored in high-dynamic-range images typically corresponds to the physical values of luminance or radiance that can be observed in the real world.',
from my perspective,HDR stores data that should be absolute values.Is my understanding correct?what dataset are you using?

Hey @LWT3437 could you please share the network architecture for LAM. (the input and output channels for the various conv layers in LAM please) I am unable to understand it from the code