Numerous style transfer methods which produce artistic styles of portraits have been proposed to date. However, the inverse problem of converting the stylized portraits back into realistic faces is yet to be investigated thoroughly. Reverting an artistic portrait to its original photo-realistic face image has potential to facilitate human perception and identity analysis. In this paper, we propose a novel Face Destylization Neural Network (FDNN) to restore the latent photo-realistic faces from the stylized ones. We develop a Style Removal Network composed of convolutional, fully-connected and deconvolutional layers. The convolutional layers are designed to extract facial components from stylized face images. Consecutively, the fully-connected layer transfers the extracted feature maps of stylized images into the corresponding feature maps of real faces and the deconvolutional layers generate real faces from the transferred feature maps. To enforce the destylized faces to be similar to authentic face images, we employ a discriminative network, which consists of convolutional and fully connected layers. We demonstrate the effectiveness of our network by conducting experiments on an extensive set of synthetic images. Furthermore, we illustrate our network can recover faces from stylized portraits and real paintings for which the stylized data was unavailable during the training phase.

Recovering the latent photorealistic faces from their artistic portraits aids human perception and facial analysis. However, a recovery process that can preserve identity is challenging because the fine details of real faces can be distorted or lost in stylized images. In this paper, we present a new Identity-preserving Face Recovery from Portraits (IFRP) to recover latent photorealistic faces from unaligned stylized portraits. Our IFRP method consists of two components: Style Removal Network (SRN) and Discriminative Network (DN). The SRN is designed to transfer feature maps of stylized images to the feature maps of the corresponding photorealistic faces. By embedding spatial transformer networks into the SRN, our method can compensate for misalignments of stylized faces automatically and output aligned realistic face images. The role of the DN is to enforce recovered faces to be similar to authentic faces. To ensure the identity preservation, we promote the recovered and ground-truth faces to share similar visual features via a distance measure which compares features of recovered and ground-truth faces extracted from a pre-trained VGG network. We evaluate our method on a large-scale synthesized dataset of real and stylized face pairs and attain state of the art results. In addition, our method can recover photorealistic faces from previously unseen stylized portraits, original paintings and human-drawn sketches.

In popular TV programs (such as CSI), a very low-resolution face image of a person, who is not even looking at the camera in many cases, is digitally super-resolved to a degree that suddenly the person's identity is made visible and recognizable. Of course, we suspect that this is merely a cinematographic special effect and such a magical transformation of a single image is not technically possible. Or, is it? In this paper, we push the boundaries of super-resolving (hallucinating to be more accurate) a tiny, non-frontal face image to understand how much of this is possible by leveraging the availability of large datasets and deep networks. To this end, we introduce a novel Transformative Adversarial Neural Network (TANN) to jointly frontalize very-low resolution (i.e. 16×16 pixels) out-of-plane rotated face images (including profile views) and aggressively super-resolve them (8×), regardless of their original poses and without using any 3D information. TANN is composed of two components: a transformative upsampling network which embodies encoding, spatial transformation and deconvolutional layers, and a discriminative network that enforces the generated high-resolution frontal faces to lie on the same manifold as real frontal face images. We evaluate our method on a large set of synthesized non-frontal face images to assess its reconstruction performance. Extensive experiments demonstrate that TANN generates both qualitatively and quantitatively superior results achieving over 4 dB improvement over the state-of-the-art.

Fig. 1 Comparisons to the state-of-art methods. (a) The ground truth face image (from test dataset, not available in the training dataset). (b) Unaligned stylized portraits of (a) from Scream style. (c) Landmarks detected by [63]. (d) Results obtained by [17]. (e) Results obtained by [64] (CycleGAN). (f) Results obtained by [14] (pix2pix). (g) Results obtained by [42]. (h) Our results.

Searching interested images based on visual properties of images is a challenging problem and it has received considerable attention from researchers in the fields like image processing, computer vision and multimedia systems in the last 20 years. While the importance and the effect of the image features like color, texture and shape have been taken into account in many papers, there have not been many studies on the importance of the color spaces on the performance of Content Based Image Retrieval (CBIR) systems. In this paper we first experimentally study the effect of choosing color space on the performance of content based image retrieval using Wavelet decomposition of each color channel. To this end, the retrieval results of different color spaces like RGB, YUV, HSV, YCbCr and Lab are analyzed. Then as a result a new Content Based Retrieval model using Wavelet Transform in Lab color space and Color Moments is proposed. In order to increase the efficiency of the proposed model some division schemes are taken into account which improves the performance of the proposed model. The proposed model tackles one of the important restrictions in content based image retrieval, namely, the challenge between the accuracy of retrieval and its time complexity. The experimental results on two databases [19] [24] demonstrate the superiority of the proposed model compared to existing models.