Vuong Ho
Vuong Ho
Hi, I think you used small dataset, As you showed, you only training 2 part per epoch
It is not work in my case, anyone can you help me @shd101wyy ``` | a | b | |---|---| | 1 | 2 | | ^ | 3 |...
Hi i think if you remove the line as above, maybe the source code can not run ```evaluate```, because they need ```predictions_all, gts_all, train_args.nb_classes``` ```acc, acc_cls, mean_iu, fwavacc, f1 =...
Hi @Dootmaan Thank you for your interest. The output of Saliency Attention Network is a probability map with a range [0,1] which is the same size as the input image....
> @hongocvuong1998 Thank you for your hard work. > > **General:** > > * What are the required dependencies & versions to run this program? > Example: (Opencv 3 or...
To fix that, you need to fix the input model is (1,1,512,512) ``` image_tensor = get_image_tensor(image_path) # shape (1, 1, 512, 512) image_tensor = image_tensor[:,0:1,:,:] ```
Please refer here https://github.com/GAP-LAB-CUHK-SZ/MVImgNet2.0/issues/4#issue-3326512591
Hi authors Thank you so much for the great publication. I am waiting for the MoGe-2 paper release!