Conference Paper

Image Inpainting Considering Brightness Change and Spatial Locality of Textures and Its Evaluation

DOI: 10.1007/978-3-540-92957-4_24 Conference: Advances in Image and Video Technology, Third Pacific Rim Symposium, PSIVT 2009, Tokyo, Japan, January 13-16, 2009. Proceedings
Source: DBLP


Image inpainting techniques have been widely investigated to remove undesired objects in an image. Conventionally, missing
parts in an image are completed by optimizing the objective function using pattern similarity. However, unnatural textures
are easily generated due to two factors: (1) available samples in the image are quite limited, and (2) pattern similarity
is one of the required conditions but is not sufficient for reproducing natural textures. In this paper, we propose a new
energy function based on the pattern similarity considering brightness changes of sample textures (for (1)) and introducing
spatial locality as an additional constraint (for (2)). The effectiveness of the proposed method is successfully demonstrated
by qualitative and quantitative evaluation. Furthermore, the evaluation methods used in much inpainting research are discussed.

16 Reads
  • Source
    • "Wexler et al. (2007) formulate the reconstruction procedure as an optimization problem which employs a combination of dynamic space–time and tree structures. Kawai et al. (2009) rely on modifications of the energy functional proposed in Wexler et al. (2007) improving the spatial localization of the similarity weights and brightness invariance. Komodakis and Tziritas (2007) employ variations of the sum-product algorithm (loopy belief propagation) for graphs with cycles associated to ''prioritybased message scheduling'' during the filling process. "
    [Show abstract] [Hide abstract]
    ABSTRACT: In this work we propose a new image inpainting technique that combines texture synthesis, anisotropic diffusion, transport equation and a new sampling mechanism designed to alleviate the computational burden of the inpainting process. Given an image to be inpainted, anisotropic diffusion is initially applied to generate a cartoon image. A block-based inpainting approach is then applied so that to combine the cartoon image and a measure based on transport equation that dictates the priority on which pixels are filled. A sampling region is then defined dynamically so as to hold the propagation of the edges towards image structures while avoiding unnecessary searches during the completion process. Finally, a cartoon-based metric is computed to measure likeness between target and candidate blocks. Experimental results and comparisons against existing techniques attest the good performance and flexibility of our technique when dealing with real and synthetic images.
    Pattern Recognition Letters 01/2014; 36(1):36–45. DOI:10.1016/j.patrec.2013.08.023 · 1.55 Impact Factor
  • Source
    • "Des variantes proposent de mélanger ces deux approches pour mettre en oeuvre des méthodes hybrides [1] [7] [12], d'autres [8] [14] de moyenner plusieurs patchs candidats avant de les recoller . Dans tous les cas, il existe toujours des configurations locales qui ne permettent pas de sélectionner des patchs satisfaisants au critère de similarité choisi. "
  • Source
    • "This is a novel feature of our method. Although, according to our previous review, the optimization of an energy functional is a common approach for exemplar-based image inpainting [31], [30], [27], [43], [34], [24], there are significant differences with our method, both in the energy definition and the optimization strategy. Let us discuss them in more detail. "
    [Show abstract] [Hide abstract]
    ABSTRACT: We present a novel formulation of exemplar-based inpainting as a global energy optimization problem, written in terms of the offset map. The proposed energy functional combines a data attachment term to ensure the continuity of the reconstruction at the boundary of the inpainting domain and a smoothness term that ensures a visually coherent reconstruction inside the hole. This formulation is adapted to obtain a global minimum using graph cuts algorithm. In order to reduce the computational complexity, we propose an efficient multiscale graph cuts algorithm. Moreover, to compensate for the loss of information at low resolution levels we use a feature representation computed at the original image resolution. This permits to alleviate the ambiguity induced by comparing only color information when the image is represented at low resolution levels. Our experiments show the good performance of the proposed algorithm when compared with other recent algorithms.
    IEEE Transactions on Image Processing 09/2012; 22(5). DOI:10.1109/TIP.2012.2218828 · 3.63 Impact Factor
Show more


16 Reads
Available from