To solve this, we propose an improved Selective Free-Form Attention (SFFA) module, selectively allowing the information flow from the valid sources to hole area with free-form shape. Meanwhile, we notice that the existing attention mechanism used to build long-distance dependency is unreasonable: either does not adapt to free-form holes, or lacks filtering for invalid information sources, which could compromise feature map reconstruction. To address this issue, we propose a novel coarse-to-fine residual inpainting framework: we first reconstruct the downsampled low-frequency coarse profile at a low computational cost, we then solely focus on generating high-frequency details, which will be added as residuals to the coarse profile, such that structure and texture details in composited result can be better preserved. However, many of the existing methods still suffer from the defects of unnatural structure and blurry textures when filling large-area holes.
#MULTIPATCH TOOL PATCH#
We believe our method could provide a new way for patch match with better accuracy and efficiency in image inpainting tasks.ĭeep learning has dominated the methodology of image inpainting in recent years. Experimental results show that in comparison to previous patch-based works, our method has achieved further improvement both in quality and efficiency. This strategy not only saves the match time for single target patch, but also reduces the mismatch, and enables the simultaneous filling of multiple target patches in a single iteration. Moreover, we divide the source region into multiple non-overlapping subregions with different nonuniformity levels, and the patch match proceeds in every subregion, respectively. To handle the issues above, we first evaluate the nonuniformity in an image, by which the patch size is adaptively determined. Also, global match is needed for searching the best sample patch, but only to fill one target patch in each iteration, resulting in low efficiency. However, most of the existing approaches basically use the fixed size of patch regardless of content features nearby, which may lead to inpainting defects. Patch-based image inpainting methods iteratively fill the missing region via searching the best sample patch from the source region.
0 kommentar(er)
