Levent Karacan

Structure-preserving image smoothing via region covariances

Recent years have witnessed the emergence of new image smoothing techniques which have provided new insights and raised new questions about the nature of this well-studied problem. Specifically, these models separate a given image into its structure and texture layers by utilizing non-gradient based definitions for edges or special measures that distinguish edges from oscillations. In this study, we propose an alternative yet simple image smoothing approach which depends on covariance matrices of simple image features, aka the region covariances. The use of second order statistics as a patch descriptor allows us to implicitly capture local structure and texture information and makes our approach particularly effective for structure extraction from texture. Our experimental results have shown that the proposed approach leads to better image decompositions as compared to the state-of-the-art methods and preserves prominent edges and shading well. Moreover, we also demonstrate the applicability of our approach on some image editing and manipulation tasks such as image abstraction, texture and detail enhancement, image composition, inverse halftoning and seam carving.

Image Matting with KL-Divergence Based Sparse Sampling

Previous sampling-based image matting methods typically rely on certain heuristics in collecting representative samples from known regions, and thus their performance deteriorates if the underlying assumptions are not satisfied. To alleviate this, in this paper we take an entirely new approach and formulate sampling as a sparse subset selection problem where we propose to pick a small set of candidate samples that best explains the unknown pixels. Moreover, we describe a new distance measure for comparing two samples which is based on KL-divergence between the distributions of features extracted in the vicinity of the samples. Using a standard benchmark dataset for image matting, we demonstrate that our approach provides more accurate results compared with the state-of-the-art methods.

Alpha Matting With KL-Divergence-Based Sparse Sampling

In this paper, we present a new sampling-based alpha matting approach for the accurate estimation of foreground and background layers of an image. Previous sampling-based methods typically rely on certain heuristics in collecting representative samples from known regions, and thus their performance deteriorates if the underlying assumptions are not satisfied. To alleviate this, we take an entirely new approach and formulate sampling as a sparse subset selection problem where we propose to pick a small set of candidate samples that best explains the unknown pixels. Moreover, we describe a new dissimilarity measure for comparing two samples which is based on KL-divergence between the distributions of features extracted in the vicinity of the samples. The proposed framework is general and could be easily extended to video matting by additionally taking temporal information into account in the sampling process. Evaluation on standard benchmark data sets for image and video matting demonstrates that our approach provides more accurate results compared with the state-of-the-art methods.

Image colorization via dense correspondences

Colorization, the process of adding color to monochrome images, is a tedious and difficult task and often requires intensive manual effort by color experts. To alleviate this problem, a number of computational studies have been proposed in the literature which aim to perform this task in a relatively easy way, either by employing minimal user input in terms of color scribbles or using a colored reference image. Our goal in this paper is to explore a fully-automatic approach to image colorization. In particular, we present a novel data-driven strategy which automatically selects the most similar reference image from a large set of color images and utilizes dense correspondences to transfer the color information from the reference image to the input image. We evaluate the performance of our approach on a variety of natural images and obtain fairly good results.

Adjusting transient attributes of outdoor images using generative adversarial networks

In our ever changing world, natural outdoor scenes undergo significant changes due to lighting, weather and seasonal conditions at different times of the day and the year. Therefore, it is remarkably challenging to build computational models which can automatically manipulate the appearance of outdoor images in a realistic manner. Suggested approaches employ several intermediate steps that may seriously affect the quality of the result, such as retrieving similar images in a large database and matching those images to the input image. As an effort to eliminate these drawbacks of the previous methods, in this paper, we present an automatic image editing approach which utilizes generative adversarial networks to learn the appearance manifold of outdoor images. Our experiments show that our model yields natural looking promising results.