Table 3 Results of different models on the Vid-QU-EX dataset.

U-Net²

0.8265

0.7992

0.7051

1.9447

245.7538

Attention-U-Net³⁸

0.8389

0.8133

0.7236

34.8786

141.7379

DCANet³⁹

0.8317

0.8051

0.7132

36.6003

31.9992

M-Net⁴⁰

0.8378

0.8123

0.7220

9.3277

189.9465

DCSAU-Net⁴¹

0.8417

0.8167

0.7280

2.5988

54.4532

MCDAU-Net⁴²

0.8351

0.8093

0.7185

12.9797

66.2665

META-Unet⁴³

0.8317

0.8050

0.7131

21.6960

86.8963

MSRAformer⁴⁴

0.7942

0.7628

0.6603

68.0315

23.1412

Swin-Transformer⁴⁵

0.7944

0.7625

0.6599

36.7198

58.9808

MCAFNet⁴⁶

0.8360

0.8099

0.7194

9.0615

81.3309

MDUNet⁴⁷

0.8315

0.8049

0.7131

11.5519

38.5120

DualA-Net⁴⁸

0.8236

0.7957

0.7013

2.5788

52.7225

MD-Net

0.8425

0.8176

0.7292

8.5747

73.0779