shuochsu / deepend2endtofimaging Goto Github PK

How to download the dataset and model? when I run the XXX.sh, there are no result and when I click the link http://www.cs.ubc.ca/labs/imager/tr/2018/DeepEnd2EndToFImaging/datasets/$FILE.zip
The webpage shows that
"The requested URL was not found on this server. If you entered the URL manually please check your spelling and try again.
Error 404
www.cs.ubc.ca
Apache/2.4.23 (Linux/SUSE)"

Thank you for uploading your work and the dataset! I have two questions:

1. Are the original pbrt EXR files available for download as well?
2. From my understanding, the DeepToF_release_0.1/scripts/pbrt_tof_batch_albedo_aug.sh script runs a pbrt-v3-tof/pbrt_all_10 command, which I cannot find. I've tried building the provided DeepToF_release_0.1/src/pbrt-v3-tof directory, but am still unable to find that command. How can I get the pbrt_all_10 command?

Thank you!

Hi, I study your paper recently, After operate: bash job_example.sh, I'll get the result, and how to start matlab part? Is Matlab program is necessary? can I modify the input and output？

How to prepare the dataset and how to get corr_img and depth_peaks?

Hello! I am working on a CNN which gets amplitude data as input (and output) and wondered how exactly you performed the normalization (value range, etc.)? I remember reading that with GANs [-1, 1] is the preferred value range which does not seem intuitive for amplitude data...

Hello, I am a student from China. Recently I was reproducing your published Deep End-to-End Time-of-Flight Imaging. when I was downloading the training dataset, the unpacking reported an error as follows.

$ unzip mpi_correction_corrnormamp2depth_albedoaug_zpass

Archive: mpi_correction_corrnormamp2depth_albedoaug_zpass.zip
warning [mpi_correction_corrnormamp2depth_albedoaug_zpass.zip]: 4294967296 extra bytes at beginning or within zipfile
(attempting to process anyway)
file #1: bad zipfile offset (local header sig): 4294967296
(attempting to re-compensate)
error: invalid zip file with overlapped components (possible zip bomb)

It may be caused by a corrupted zip file, can you please provide a normal zip file?

Nice work! When I used your code, I found that the pre-training model download link in the code has failed. Could you please upload the pre-training models? Thanks a lot.

Hello,

I studied your very interesting paper and found that you use a depth smoothing loss like in https://arxiv.org/abs/1609.03677.

However, the cited work used disparity (basically the inverse of depth). Did you use the actual depth or the inverse?

Thank you very much

Hello, can you still use your code to improve the MPI problem?
There is very little open source code about this, I need your help Thank you

Hello! I'm from China and I'm researching the problem of multipath of ToF using your method now. And when I use your net structure and the synthetic data for training model, I find the training speed is very slow, more slower than you said in your article(my machine: ubuntu16.04 64bit, GPU GTX1080 8GB, i7-7700 CPU, 16GB RAM, I use CUDA and cuDNN), maybe exceed 1,000 hours, So I want to know is the Net parameter: batchSize is 1? Or other parameters are changing? Please, Thank you a lot.