svishwa / crowdcount-mcnn

Single Image Crowd Counting via MCNN (Unofficial Implementation)
MIT License
504 stars 180 forks source link

How to get the './data/original/shanghaitech/part_B_final/test_data/ground_truth_csv/' #31

Open Xuwanjun9 opened 6 years ago

Xuwanjun9 commented 6 years ago

This document './data/original/shanghaitech/part_B_final/test_data/ground_truth_csv/' is used in test.py. But the dataset I got doesn't contain this document. Can anyone tell me how can I get this document? Thanks

BlakeRainey commented 5 years ago

You need to run create_gt_test_set_shtech.m

Xuwanjun9 commented 5 years ago

You need to run create_gt_test_set_shtech.m

I did, but there's no such a document too. I got the ground_truth_csv document in part_A only, but no such a document in part_B

Xuwanjun9 commented 5 years ago

You need to run create_gt_test_set_shtech.m

I found the reason why I failed before. Thank you so much for your answer, thanks!

ghost commented 5 years ago

@Xuwanjun9 same problem here. Can you tell me the resolvance here. I am getting error like below - "error: load: unable to find file ../data/original/shanghaitech/part_A_final/test_data/ground_truth/GT_IMG_1.mat error: called from create_gt_test_set_shtech at line 24 column 5"

Xuwanjun9 commented 5 years ago

@Xuwanjun9 same problem here. Can you tell me the resolvance here. I am getting error like below - "error: load: unable to find file ../data/original/shanghaitech/part_A_final/test_data/ground_truth/GT_IMG_1.mat error: called from create_gt_test_set_shtech at line 24 column 5" you need run the matlab file first

ghost commented 5 years ago

The error is from matlab console only, When I ran the create_gt_test_set_shtech.m file.

BlakeRainey commented 5 years ago

As I remember there were underscores used in the MATLAB file refering to the dataset but the dataset itself uses hyphens. That's probably why the file can't be found

koalaofpoint commented 5 years ago

@Xuwanjun9 Sir ,I found ./data/original/shanghaitech/part_B_final/test_data/ground_truth_csv/ is not exists,I had run……m file ,Could you tell me the reason?

Xuwanjun9 commented 5 years ago

@Xuwanjun9 Sir ,I found ./data/original/shanghaitech/part_B_final/test_data/ground_truth_csv/ is not exists,I had run……m file ,Could you tell me the reason?

open the .m file, you'll see that there's a para indicate the dataset A or B only, change it and run again

koalaofpoint commented 5 years ago

@Xuwanjun9 Sir,Thanks for your reply,I have solve the question.But a new question is occur that I cann't understand the create_gt_train_set_shtech.m,I have scan the web without solve the question,can you help me

Xuwanjun9 commented 5 years ago

@Xuwanjun9 Sir,Thanks for your reply,I have solve the question.But a new question is occur that I cann't understand the create_gt_train_set_shtech.m,I have scan the web without solve the question,can you help me

What makes you confused? Have you open and read the .m file?

koalaofpoint commented 5 years ago

@Xuwanjun9 I have open the .m file ,and the code have no antotation,many of that I cann't understand,so I want a detail file but the brower haven't ,so your code hava antotation?

jinryan commented 5 years ago

@Xuwanjun9 @koalaofpoint in the create_gt_test_set_shtech.m file, change the dataset='A' to dataset='B' and you will be able to generate the ground truth csv for part B. I struggled with the same problem until I looked at the code.

tongpinmo commented 5 years ago

@Xuwanjun9 do you know how to get the GT_x.mat file for own dataset. the structure is so stange ,like ('image_info: ', array([[array([[(array([[222.5446098 , 472.86475499], [ 6.63269208, 709.19046582], [ 64.82506957, 694.96566243], ..., [714.76278388, 418.8815984 ], [519.37417249, 234.74573427], [523.46608059, 238.83764236]]), array([[707]], dtype=uint16))]], dtype=[('location', 'O'), ('number', 'O')])]], dtype=object))

is unlike python version,i am confused

Xuwanjun9 commented 4 years ago

@Xuwanjun9 do you know how to get the GT_x.mat file for own dataset. the structure is so stange ,like ('image_info: ', array([[array([[(array([[222.5446098 , 472.86475499], [ 6.63269208, 709.19046582], [ 64.82506957, 694.96566243], ..., [714.76278388, 418.8815984 ], [519.37417249, 234.74573427], [523.46608059, 238.83764236]]), array([[707]], dtype=uint16))]], dtype=[('location', 'O'), ('number', 'O')])]], dtype=object))

is unlike python version,i am confused

It's a matlab code, you have to run this code in matlab to get ground truth

tongpinmo commented 4 years ago

@Xuwanjun9 do you know how to get the GT_x.mat file for own dataset. the structure is so stange ,like ('image_info: ', array([[array([[(array([[222.5446098 , 472.86475499], [ 6.63269208, 709.19046582], [ 64.82506957, 694.96566243], ..., [714.76278388, 418.8815984 ], [519.37417249, 234.74573427], [523.46608059, 238.83764236]]), array([[707]], dtype=uint16))]], dtype=[('location', 'O'), ('number', 'O')])]], dtype=object)) is unlike python version,i am confused

It's a matlab code, you have to run this code in matlab to get ground truth

ok ,I have solved it ,thank you