programminglearner / matf-architecture-details Goto Github PK

Hi
I'm interesting in your paper and work. Did you train the model when encountering different data set? Or you train a general model that can get good result on all data set?
Thanks

up3 = F.upsample(conv3, scale_factor=5)
features = conv1 + up2 + up3
up3 is not the same dimension with conv1 and up2, it seems that the scale_factor is error

your height and width are 30. In this case, this is right

Hi, thanks for the excellent work! May I ask about the detailed configuration regarding SDD dataset preprocessing? For example, did you use the raw SDD data downloaded from https://cvgl.stanford.edu/projects/uav_data/? If yes, how did you filter the dirty data(e.g. lost, unstable camera) and decide the volume of your training/val/test set? Also, did you do any data augmentation on it? Any help will be appreciated!

Hi!

Can you share the datasets that you used to train the model?

Best regards,
Bruno

Hi,

can you opensource code to train and validate model>?