Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Inquire the meaning of data #2

Closed
yangengt123 opened this issue May 16, 2019 · 12 comments
Closed

Inquire the meaning of data #2

yangengt123 opened this issue May 16, 2019 · 12 comments
Labels
question Further information is requested

Comments

@yangengt123
Copy link

yangengt123 commented May 16, 2019

Hi,
Thanks for sharing the code of your great work.

I am trying to run the code to get a better insight of your paper, but I get confused about the variable meanings in the input_dict. In the dict, I assume
lmap: line heatmap, but may I ask what the float value means? I thought it should be binary
jmap: junction heatmap
joff: junction offset map,

but I cannot understand what junc, Lpos, Lneg , lpos, and lneg are. Especially, I found the the shape of Lpos, Lneg is n*2, while lpos and lneg has a shape n*2*3.

Could you offer more explanation about them?

Thanks in advance!

@zhou13 zhou13 added the question Further information is requested label May 16, 2019
@zhou13
Copy link
Owner

zhou13 commented May 16, 2019

See dataset/wireframe.py to understand the details about data pre-processing.

@yangengt123
Copy link
Author

yangengt123 commented May 16, 2019

See dataset/wireframe.py to understand the details about data pre-processing.

Thanks for the quick reply.
Which part are you indicating in dataset/wireframe.py? Actually, I have read both this file and the .json in raw data .tar, but I did not get any cue about the physical meaning of them.

@zhou13
Copy link
Owner

zhou13 commented May 16, 2019

lmap: line heatmap, but may I ask what the float value means? I thought it should be binary

rr, cc, value = skimage.draw.line_aa(*vint0, *vint1)

which part are you indicating in dataset/wireframe.py?

lcnn/dataset/wireframe.py

Lines 99 to 103 in 36f172d

junc = np.array(junc, dtype=np.float32)
Lpos = np.array(lnid, dtype=np.int)
Lneg = np.array([l[2:4] for l in lneg][:4000], dtype=np.int)
lpos = np.array(lpos, dtype=np.float32)
lneg = np.array([l[:2] for l in lneg[:2000]], dtype=np.float32)

@zhou13 zhou13 closed this as completed in 7aee92c May 16, 2019
@zhou13
Copy link
Owner

zhou13 commented May 16, 2019

I add more comments to dataset/wireframe.py and lcnn/datasets.py to explain the meaning of each variable you asked. Hopefully, it is more clear now.

@yangengt123
Copy link
Author

Thanks. It helps.
BTW may I ask if there is any implementation for test on image data only?
I plan to perform test on the my own dataset, which contains only the RGB image. I found the code for test (process.py) seems still need meta data to run. I am trying to modify the code for my test, but it will be perfect if there is any existing test code.

@zhou13
Copy link
Owner

zhou13 commented May 17, 2019

Not yet as I am busy with a paper submission. You are more than welcome to contribute that if you finish it :)

@sachinkaundal
Copy link

@zhou13, can you please tell what kind of data train.json and valid.json consist of that you shared with dataset and how you generated. Thanks in advance for response.

@zhou13
Copy link
Owner

zhou13 commented Jul 7, 2021

@sachinkaundal Not sure what kind of answer you need. The code at lcnn/dataset.py should explain your question well.

@sachinkaundal
Copy link

sachinkaundal commented Jul 8, 2021

@zhou13 Thanks for the response. i am not able to understand train.json and valid.json(shared with wireframe_raw) files that are used during processing(wireframe.py). Even i read wireframe.py file i am not getting any cue about these files.
Thanks

@zhou13
Copy link
Owner

zhou13 commented Jul 8, 2021

You might want to see this repo for the dataset: https://github.com/huangkuns/wireframe.

@sachinkaundal
Copy link

@zhou13, Thank you so much for your time.

@SEUZTh
Copy link

SEUZTh commented Mar 11, 2022

Hello, I have downloaded the dataset from Onedrive. It seems that the dataset's directory structure in Onedrive is different from that in the compressed package. Could you share the directory structure with me? I'm looking forward to your reply.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
question Further information is requested
Projects
None yet
Development

No branches or pull requests

4 participants