search

svishwa / crowdcount-mcnn

Single Image Crowd Counting via MCNN (Unofficial Implementation)
MIT License
499 stars 179 forks source link

Dataset required #11

Closed asif07hossain closed 6 years ago

asif07hossain commented 6 years ago

Dear, You previously helped me with sending the Shanghaitech dataset. Now, I am need of Wroldexpo'10 dataset and UCSD dataset.

Let me explain you a scenario about UCF_CC_50 dataset. I have 50 images. Now I put first 40 images in to train folder like shanghaitech, then rest 10 images i put on test folder. finally i got 359 formatted images to train my model. I want to know how authors tested their method with UCF_CC_50 dataset.

Can you please reply me with these 3 queries. Thanks in advance. asif07hossain@gmail.com is my gmail id

asif07hossain commented 6 years ago

Dear Concern,

Can you please provide me the UCF_CC_50, UCSD and Wroldexpo'10 dataset pre-processing codes to divide test and training? my gmail id is: asif07hossain@gmail.com

svishwa commented 6 years ago

The pre-processing code provided for ShanghaiTech dataset can be modified for other datasets. Please follow the cross-validation split as discussed in the UCF_CC_50 dataset paper to perform the experiments.

xxxwuwq commented 4 years ago

The pre-processing code provided for ShanghaiTech dataset can be modified for other datasets. Please follow the cross-validation split as discussed in the UCF_CC_50 dataset paper to perform the experiments.

the description about 5-fold cross-validation in paper, not specified the times about multiple 5-fold cross-validation. As for current paper, it's unfair to compare the result on ucf_cc_50 for that there is no uniform standard(such as 5 times 5-fold cross validation, instead of once specified combination(handcraft))???