Test/metric_evaluation.py

import os
import cv2
import numpy as np
from evaluate import comput_sad_loss, compute_mse_loss, compute_connectivity_error, compute_gradient_loss 
import argparse


if __name__ == '__main__':
    parser = argparse.ArgumentParser()
    parser.add_argument('--pred-dir', type=str, default='./predDIM/', help="pred alpha dir")
    parser.add_argument('--label-dir', type=str, default='./Test_set/alpha_copy/', help="GT alpha dir")
    parser.add_argument('--trimap-dir', type=str, default='./Test_set/trimaps/', help="trimap dir")

    args = parser.parse_args()

    mse_loss = []
    sad_loss = []

    ### loss_unknown only consider the unknown regions, i.e. trimap==128, as trimap-based methods do
    mse_loss_unknown = []
    sad_loss_unknown = []

    grad_loss = []
    compute_connectivity_loss = []

    grad_loss_unknown = []
    compute_connectivity_loss_unknown = []
 
    for img in os.listdir(args.pred_dir):
        label = cv2.imread(os.path.join(args.label_dir, img), 0).astype(np.float32)
        pred = cv2.imread(os.path.join(args.pred_dir, img), 0).astype(np.float32)
        trimap = cv2.imread(os.path.join(args.trimap_dir, img), 0).astype(np.float32)
        if pred.shape != label.shape:
            pred = cv2.resize(pred, (label.shape[1], label.shape[0]))

        mse_loss_unknown_ = compute_mse_loss(pred, label, trimap)
        sad_loss_unknown_ = comput_sad_loss(pred, label, trimap)[0]
        gradient_loss_unknown = compute_gradient_loss(pred, label, trimap) 
        connectivity_loss_unknown = compute_connectivity_error(pred, label, trimap, 0.1)

        trimap[...] = 128
        mse_loss_ = compute_mse_loss(pred, label, trimap)
        sad_loss_ = comput_sad_loss(pred, label, trimap)[0]
        gradient_loss = compute_gradient_loss(pred, label, trimap)
        connectivity_loss = compute_connectivity_error(pred, label, trimap, 0.1)

        print('Whole Image: MSE: ', mse_loss_, ' SAD:', sad_loss_, "GRAD:", gradient_loss, "Conn:", connectivity_loss)
        print('Unknown Region: MSE:', mse_loss_unknown_, ' SAD:', sad_loss_unknown_, "GRAD:", gradient_loss_unknown, "Conn:", connectivity_loss_unknown)


        mse_loss_unknown.append(mse_loss_unknown_)
        sad_loss_unknown.append(sad_loss_unknown_)

        mse_loss.append(mse_loss_)
        sad_loss.append(sad_loss_)

        grad_loss.append(gradient_loss)
        compute_connectivity_loss.append(connectivity_loss)

        grad_loss_unknown.append(gradient_loss_unknown)
        compute_connectivity_loss_unknown.append(connectivity_loss_unknown)

    print('Average:')
    print('Whole Image: MSE:', np.array(mse_loss).mean(), ' SAD:', np.array(sad_loss).mean())
    print('Unknown Region: MSE:', np.array(mse_loss_unknown).mean(), ' SAD:', np.array(sad_loss_unknown).mean())
    print("whole Grad, Conn:", np.array(grad_loss).mean(), ' compute_connectivity_loss:', np.array(compute_connectivity_loss).mean())
    print("Unknown GRAD, CONN:", np.array(grad_loss_unknown).mean(), ' CONN:', np.array(compute_connectivity_loss_unknown).mean())