I find it difficult to conduct operations between fields. In my case, a sample consists of two images: A and B. I have to ensure the consistency of the operations on the two images. For instance, I can easily implement it on the PyTorch style dataset:
i, j, h, w = get_crop_c(height, width, scale, ratio)
resize_crop_c(temp_buffer, i, i + h, j, j + w,
A)
resize_crop_c(temp_buffer, i, i + h, j, j + w,
B)
Is it possible to use the same random crop on two fields?
I find it difficult to conduct operations between fields. In my case, a sample consists of two images: A and B. I have to ensure the consistency of the operations on the two images. For instance, I can easily implement it on the PyTorch style dataset:
Is it possible to use the same random crop on two fields?