haqishen / mfnet-pytorch Goto Github PK

I tried to print them but it seems the values are wrong, could anyone tell me the right value for each class?

When trying to replicate your findings I noticed an odd detail about the calculation of the mIoU. It concerns the formula (2) in the paper.
The outer sum, as claimed in the text, iterates over i=2..n in order to compute the average IoU over all classes except background (which corresponds to i=1). My issue is with the inner sum, also iterating over j=2..n. This does not seem correct to me as it effectively disregards all pixels that are marked as background by either the ground truth or the model. In the extreme case, this would mean that a class, where only one pixel of ground truth and model output overlaps, is awarded an IoU of 1.0 if all the remaining pixels are background.

To measure the error incurred, I used the model weights saved in the repository to redo the mIoU calculation, and obtained a value of 0.290 instead of 0.649, as claimed in the paper.

Is there a specific reason for this way of calculating the IoU and/or a reference you could point me towards?

I'm confused about conf[:,cid] = cf[:,cid]/cf[:,cid].sum() in calculate_result() function(util.py).
In this paper and other papers, the definition of class accuracy is: $\frac{P_{ii}}{\sum_{j} P_{ij}$,
in this way, the code should be conf[cid,:] = cf[cid,:]/cf[cid,:].sum()

Excuse me,I am very interested in the data set of this paper. But I failed to download the dataset. If you are free, can you send it to my email? I would appreciate it. My emails: [email protected]

Please give access for dataset, the website link in readme file is not working

Access permission is demanded now for dataset downloading on Google drive. Could you provide that plz?

I wonder which method you used to label the image as your training ground truth. Labeled the Thermal and RGB images separately, and then intergrated them? Or use some other methods?

Thanks for sharing your code here!

Environment: Anaconda(python 2.7x)+Pytorch0.4.1
Problem: Continuous error occurs when trying to run the demo

It seems to be python 3.x considering the print function you use in your code, is that right?

Hello. I just took a look at your great work and played around with your dataset and scripts and was wondering how you labeled your dataset? Is it in the yolo-style?

I want to ask about the dataset 'images'. I find that the input images are all four channels and I want to know how it is made from the rgb input and the thermal input.