Image Matting: A Comprehensive Survey on Techniques, Comparative Analysis, Applications and Future Scope

In the era of rapid growth of technologies, image matting plays a key role in image and video editing along with image composition. In many significant real-world applications such as film production, it has been widely used for visual effects, virtual zoom, image translation, image editing and video editing. With recent advancements in digital cameras, both professionals and consumers have become increasingly involved in matting techniques to facilitate image editing activities. Image matting plays an important role to estimate alpha matte in the unknown region to distinguish foreground from the background region of an image using an input image and the corresponding trimap of an image which represents a foreground and unknown region. Numerous image matting techniques have been proposed recently to extract high-quality matte from image and video sequences. This paper illustrates a systematic overview of the current image and video matting techniques mostly emphasis on the current and advanced algorithms proposed recently. In general, image matting techniques have been categorized according to their underlying approaches, namely, sampling-based, propagation-based, combination of sampling and propagation-based and deep learning-based algorithms. The traditional image matting algorithms depend primarily on color information to predict alpha matte such as sampling-based, propagation-based or combination of sampling and propagation-based algorithms. However, these techniques mostly use low-level features and suffer from high-level background which tends to produce unwanted artifacts when color is same or semi-transparent in the foreground object. Image matting techniques based on deep learning have recently introduced to address the shortcomings of traditional algorithms. Rather than simply depending on the color information, it uses deep learning mechanism to estimate the alpha matte using an input image and the trimap of an image. A comprehensive survey on recent image matting algorithms and in-depth comparative analysis of these algorithms has been thoroughly discussed in this paper.

Akuisisi Foreground dan Background Berbasis Fitur DTC pada Matting Citra secara Otomatis

<p>Teknik pemisahan <em>foreground</em> dari <em>background</em> pada citra statis merupakan penelitian yang sangat diperlukan dalam <em>computer vision</em>. Teknik yang sering digunakan adalah <em>image segmentation,</em> namun hasil ekstraksinya masih kurang akurat. <em>Image matting</em> menjadi salah satu solusi untuk memperbaiki hasil dari <em>image segmentation</em>. Pada metode <em>supervised</em>, <em>image matting</em> membutuhkan <em>scribbles</em> atau <em>trimap</em> sebagai <em>constraint</em> yang berfungsi untuk melabeli daerah tersebut adalah <em>foreground</em> atau <em>background</em>. Pada makalah ini dibangun metode <em>unsupervised</em> dengan mengakuisisi <em>foreground</em> dan <em>background</em> sebagai <em>constraint</em> secara otomatis. Akuisisi <em>background</em> ditentukan dari varian nilai fitur DCT (<em>Discrete Cosinus Transform</em>) yang dikelompokkan menggunakan algoritme <em>k-means</em>. Untuk mengakuisisi <em>foreground</em> ditentukan dari subset hasil klaster fitur DCT dengan fitur <em>edge detection.</em> Hasil dari proses akuisisi <em>foreground</em> dan <em>background</em> tersebut dijadikan sebagai <em>constraint</em>. Perbedaan hasil dari penelitian diukur menggunakan MAE (<em>Mean Absolute Error</em>) dibandingkan dengan metode <em>supervised matting</em> maupun dengan metode <em>unsupervised matting</em> lainnya. Skor MAE dari hasil eksperimen menunjukkan bahwa nilai <em>alpha matte</em> yang dihasilkan mempunyai perbedaan 0,0336 serta selisih waktu proses 0,4 detik dibandingkan metode <em>supervised matting</em>. Seluruh data citra berasal dari citra yang telah digunakan para peneliti sebelumnya</p><p><em><strong>Abstract</strong></em></p><p class="Abstract"><em>The technique of separating the foreground and the background from a still image is widely used in computer vision. Current research in this technique is image segmentation. However, the result of its extraction is considered inaccurate. Furthermore, image matting is one solution to improve the effect of image segmentation. Mostly, the matting process used scribbles or trimap as a constraint, which is done manually as called a supervised method. The contribution offered in this paper lies in the acquisition of foreground and background that will be used to build constraints automatically. Background acquisition is determined from the variant value of the DCT feature that is clustered using the k-means algorithm. Foreground acquisition is determined by a subset resulting from clustering DCT values with edge detection features. The results of the two stages will be used as an automatic constraint method. The success of the proposed method, the constraint will be used in the supervised matting method. The difference in results from In the research experiment was measured using MAE (Mean Absolute Error) compared with the supervised matting method and with other unsupervised matting methods. The MAE score from the experimental results shows that the alpha matte value produced has a difference of 0.336, and the difference in processing time is 0.4 seconds compared to the supervised matting method. All image data comes from images that have been used by previous researchers.</em><strong></strong></p><p><em><strong><br /></strong></em></p>

An Active Learning Framework for Alpha Matting

Good trimap is essential for high-quality alpha matte. However, making high-quality trimap is hardwork, especially for complex images. In this paper, an active learning framework is proposed to make high quality trimap. There are two active learning methods which are employed: minimization of uncertainty sampling (MUS) and maximization of expected model output change (EMOC). MUS model finds the informative area in image which can decrease the uncertain sampling of alpha matte. EMOC model finds the important areas in image which can give the maximum expected output change of alpha matte. Two methods are combined to define the active map. Active map shows important areas which are informative in image. It can help users to make high quality trimap. The analysis and evaluation of benchmark datasets show that proposed method is effective.

Deep Propagation Based Image Matting

In this paper, we propose a deep propagation based image matting framework by introducing deep learning into learning an alpha matte propagation principal. Our deep learning architecture is a concatenation of a deep feature extraction module, an affinity learning module and a matte propagation module. These three modules are all differentiable and can be optimized jointly via an end-to-end training process. Our framework results in a semantic-level pairwise similarity of pixels for propagation by learning deep image representations adapted to matte propagation. It combines the power of deep learning and matte propagation and can therefore surpass prior state-of-the-art matting techniques in terms of both accuracy and training complexity, as validated by our experimental results from 243K images created based on two benchmark matting databases.

SALIENCY-BASED UNSUPERVISED IMAGE MATTING

Spectral matting is the state-of-the-art matting method and can well solve the highly under-conditioned matte problem without manual intervention. However, it suffers from huge computation cost and inaccurate alpha matte. This paper presents a modified spectral matting method which combines saliency detection algorithm to get a higher accuracy of alpha matte with less computational cost. First, the saliency detection algorithm is used to detect general locations of foreground objects. For saliency detection method, original two-stage scheme is replaced by feedback scheme to get a more suitable saliency map for unsupervised image matting. Next, matting components are obtained through a linear transformation of the smallest eigenvectors of the matting Laplacian matrix. Then, the improved saliency map is used for grouping matting components. Finally, the alpha matte is obtained based on matte cost function. Experiments show that the proposed method outperforms the state-of-the-art methods based on spectral matting both in speed and alpha matte accuracy.