Figure 11
From: Image inpainting based on sparse representations with a perceptual metric
Second set of original, corrupted, and resulting images obtained by the proposed method. Top row: original images, middle row: corrupted images including missing areas, and bottom row: results obtained by the proposed method. The sizes of images 8 to 11 are 480 × 360, 640 × 480, 640 × 480, and 640 × 480 pixels, respectively. The percentages of missing areas are 11.3%, 5.5%, 6.2%, and 6.2% in images 8 to 11, respectively.