1. What contributions have the authors mentioned in the paper "Gradient based image completion by solving the poisson equation" ?
This paper presents a novel gradient-based image completion algorithm for removing significant objects from natural images or photographs.. A new patch-matching criterion is developed in their approach to govern the completion of gradient maps.. Several examples and comparisons are given at the end of the paper to demonstrate the performance of their gradient-based image completion approach.
read more
2. What are the future works mentioned in the paper "Gradient based image completion by solving the poisson equation" ?
Currently, the authors are intending to extend their approach from the still photography completion to the areas of video completion and the completion of mesh surfaces ( e. g., [ 20 ] ).
read more
![Fig. 4. Illustration for showing the performance of our patch-matching criterion and the gradient-based approach: (a) the completion by [5] but using our new patch– matching criterion; (b) the result from our gradient-based algorithm but with the SSD criterion; (c) the result of our full approach.](/figures/fig-4-illustration-for-showing-the-performance-of-our-patch-1bd649nl.webp)
![Fig. 5. Example IV: (a) a photo of the oil painting “Still Life with Apples”, by P. Cézanne, c. 1890; (b) the manually selected region to be completed; (c) result obtained by [6]; (d) result obtained by our gradient method. Note that our result is more sharp.](/figures/fig-5-example-iv-a-a-photo-of-the-oil-painting-still-life-2u9iam7k.webp)
![Fig. 1. Example I: (a) original image; (b) the figure needs to be completed (in white with red boundary); (c), (d) the initial gradient maps in horizontal and vertical directions; (e), (f) the inpainted gradient maps in horizontal and vertical directions; (g) the result of [6]; (h) the result by our method.](/figures/fig-1-example-i-a-original-image-b-the-figure-needs-to-be-29gbtkyq.webp)
