Elon
Elon
did someone achieve the results shown in the paper?
@0bserver07 maybe use the resnet, it will works better?
hi, I have the same problem,such as how to convert my own or the ADE2016 dataset to the coco training format.
同问,项目看起来很久没动了
I tried grid_sample and it works for me as follows: ``` def image_warp(img, depth, padding_mode='zeros'): # img: the source image (where to sample pixels) -- [B, 3, H, W] #...
@alwynmathew according to my understanding, pixel_coords need to be the shape of # [B, H, W, 3], that Z_norm is added into just like X_norm and Y_norm. I did not...