rainsun1

The training data is not needed and you only have to provide an input RGB image, a trimap (320x320) and a trained model. You can modify the corresponding codes in...

perhaps you can modify the input shape from (320, 320, 4) to (None, None, 4) in test?

@ahsanbarkati Where does the error come from? Is it in the shape computation from a layer?

You can also construct the model with a large shape such as: image_size = (800, 800) input_shape = image_size + (4, ) model = build_encoder_decoder(shapeInput=input_shape) after model prediction, you can...

1. T-Net pre-train: the model shoule only include the t-net part, and you can only train the model with L_t loss. 2. M-Net pretrain: the model shoule only include the...

You should crop the image into size (256, 256) by "image, trimap, alpha = random_scale_and_creat_patch(image, trimap, alpha, patch_size)" as in the code. Perhaps your image size is 800x600, and 600...

@PanJinquan could you ALSO send me your pre-trained model and the training datasets? Thanks! email: [email protected]