Figure 1: Given a reference style image (a) and an input image (b), we seek to create an output image of the same scene as the input, but with the style of the reference image. The Neural Style algorithm [5] (c) successfully transfers colors, but also introduces distortions that make the output look like a painting, which is undesirable in the context of photo style transfer. In comparison, our result (d) transfers the color of the reference style image equally well while preserving the photorealism of the output. On the right (e), we show 3 insets of (b), (c), and (d) (in that order). Zoom in to compare results. AbstractThis paper introduces a deep-learning approach to photographic style transfer that handles a large variety of image content while faithfully transferring the reference style. Our approach builds upon the recent work on painterly transfer that separates style from the content of an image by considering different layers of a neural network. However, as is, this approach is not suitable for photorealistic style transfer. Even when both the input and reference images are photographs, the output still exhibits distortions reminiscent of a painting. Our contribution is to constrain the transformation from the input to the output to be locally affine in colorspace, and to express this constraint as a custom fully differentiable energy term. We show that this approach successfully suppresses distortion and yields satisfying photorealistic style transfers in a broad variety of scenarios, including transfer of the time of day, weather, season, and artistic edits.

Intrinsic image decomposition separates an image into a reflectance layer and a shading layer. Automatic intrinsic image decomposition remains a significant challenge, particularly for real-world scenes. Advances on this longstanding problem have been spurred by public datasets of ground truth data, such as the MIT Intrinsic Images dataset. However, the difficulty of acquiring ground truth data has meant that such datasets cover a small range of materials and objects. In contrast, real-world scenes contain a rich range of shapes and materials, lit by complex illumination. In this paper we introduce Intrinsic Images in the Wild , a large-scale, public dataset for evaluating intrinsic image decompositions of indoor scenes. We create this benchmark through millions of crowdsourced annotations of relative comparisons of material properties at pairs of points in each scene. Crowdsourcing enables a scalable approach to acquiring a large database, and uses the ability of humans to judge material comparisons, despite variations in illumination. Given our database, we develop a dense CRF-based intrinsic image algorithm for images in the wild that outperforms a range of state-of-the-art intrinsic image algorithms. Intrinsic image decomposition remains a challenging problem; we release our code and database publicly to support future research on this problem, available online at http://intrinsic.cs.cornell.edu/.

Recognizing materials in real-world images is a challenging task. Real-world materials have rich surface texture, geometry, lighting conditions, and clutter, which combine to make the problem particularly difficult. In this paper, we introduce a new, large-scale, open dataset of materials in the wild, the Materials in Context Database (MINC), and combine this dataset with deep learning to achieve material recognition and segmentation of images in the wild.MINC is an order of magnitude larger than previous material databases, while being more diverse and well-sampled across its 23 categories. Using MINC, we train convolutional neural networks (CNNs) for two tasks: classifying materials from patches, and simultaneous material recognition and segmentation in full images. For patch-based classification on MINC we found that the best performing CNN architectures can achieve 85.2% mean class accuracy. We convert these trained CNN classifiers into an efficient fully convolutional framework combined with a fully connected conditional random field (CRF) to predict the material at every pixel in an image, achieving 73.1% mean class accuracy. Our experiments demonstrate that having a large, well-sampled dataset such as MINC is crucial for real-world material recognition and segmentation.

a) Query 1: Input scene and box (b) Project into 256D embedding (c) Results 2: use of product in-situ Convolutional Neural Network Learned Parameters θ (a) Query 2: Product (c) Results 1: visually similar products Figure 1: Visual search using a learned embedding. Query 1: given an input box in a photo (a), we crop and project into an embedding (b) using a trained convolutional neural network (CNN) and return the most visually similar products (c). Query 2: we apply the same method to search for in-situ examples of a product in designer photographs. The CNN is trained from pairs of internet images, and the boxes are collected using crowdsourcing. The 256D embedding is visualized in 2D with t-SNE. Photo credit: Crisp Architects and Rob Karosis (photographer). AbstractPopular sites like Houzz, Pinterest, and LikeThatDecor, have communities of users helping each other answer questions about products in images. In this paper we learn an embedding for visual search in interior design. Our embedding contains two different domains of product images: products cropped from internet scenes, and products in their iconic form. With such a multi-domain embedding, we demonstrate several applications of visual search including identifying products in scenes and finding stylistically similar products. To obtain the embedding, we train a convolutional neural network on pairs of images. We explore several training architectures including re-purposing object classifiers, using siamese networks, and using multitask learning. We evaluate our search quantitatively and qualitatively and demonstrate high quality results for search across multiple visual domains, enabling new applications in interior design.

We propose Deep Feature Interpolation (DFI), a new datadriven baseline for automatic high-resolution image transformation. As the name suggests, DFI relies only on simple linear interpolation of deep convolutional features from pre-trained convnets. We show that despite its simplicity, DFI can perform high-level semantic transformations like "make older/younger", "make bespectacled", "add smile", among others, surprisingly well-sometimes even matching or outperforming the state-of-the-art. This is particularly unexpected as DFI requires no specialized network architecture or even any deep network to be trained for these tasks. DFI therefore can be used as a new baseline to evaluate more complex algorithms and provides a practical answer to the question of which image transformation tasks are still challenging after the advent of deep learning.