dk-liang / TransCrowd

[SCIS] TransCrowd: Weakly-Supervised Crowd Counting with Transformers

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

AssertionError: Input image size (768*1152) doesn't match model (384*384)

YuYue26 opened this issue · comments

Input image size ({H}{W}) doesn't match model ({self.img_size[0]}{self.img_size[1]})

想复现代码,但是报了上面的错误,不知道什么原因

We divide each 758 x 1152 into six independent sub-images with 384 x 384. You can utilize the predataset_xx.py to generate the data.