Figure 5From: Image inpainting based on sparse representations with a perceptual metric Obtained results of image 2. (a) Original image (480 × 359 pixels, 24-bit color levels), (b) corrupted image including text regions (8.9% loss), (c) results obtained by the proposed method, (d) results obtained by the method in[31], (e) results obtained by the method in[34], (f) results obtained by the method in[35], (g) results obtained by the method in[24], (h) results obtained by the method in[30], and (i) results obtained by the method in[41].Back to article page