Create GT for a different dataset

Pongpisit-Thanasutives / Variations-of-SFANet-for-Crowd-Counting

The official implementation of "Encoder-Decoder Based Convolutional Neural Networks with Multi-Scale-Aware Modules for Crowd Counting"

https://ieeexplore.ieee.org/document/9413286

GNU General Public License v3.0

110 stars 32 forks source link

Create GT for a different dataset #27

Open javiercalle97 opened 2 years ago

javiercalle97 commented 2 years ago

I'm trying to understand how can I get the third column of the .npy training file.

What's the meaning of that value? Is there any script to generate it?

Pongpisit-Thanasutives commented 2 years ago

If you aim to generate the density map ground truth for each image, take a look at SFANet. To prepare for the Bayesian crowd counting, see bayesian_preprocess_sh.py. I hope this helps.