Edgar Simo-Serra

Discriminative Learning of Deep Convolutional Feature Point Descriptors

Deep learning has revolutionalized image-level tasks such as classification, but patch-level tasks, such as correspondence, still rely on hand-crafted features, e.g. SIFT. In this paper we use Convolutional Neural Networks (CNNs) to learn discriminant patch representations and in particular train a Siamese network with pairs of (non-)corresponding patches. We deal with the large number of potential pairs with the combination of a stochastic sampling of the training set and an aggressive mining strategy biased towards patches that are hard to classify. By using the L2 distance during both training and testing we develop 128-D descriptors whose euclidean distances reflect patch similarity, and which can be used as a drop-in replacement for any task involving SIFT. We demonstrate consistent performance gains over the state of the art, and generalize well against scaling and rotation, perspective transformation, non-rigid deformation, and illumination changes. Our descriptors are efficient to compute and amenable to modern GPUs, and are publicly available.

Let there be color!: joint end-to-end learning of global and local image priors for automatic image colorization with simultaneous classification

We present a novel technique to automatically colorize grayscale images that combines both global priors and local image features. Based on Convolutional Neural Networks, our deep network features a fusion layer that allows us to elegantly merge local information dependent on small image patches with global priors computed using the entire image. The entire framework, including the global and local priors as well as the colorization model, is trained in an end-to-end fashion. Furthermore, our architecture can process images of any resolution, unlike most existing approaches based on CNN. We leverage an existing large-scale scene classification database to train our model, exploiting the class labels of the dataset to more efficiently and discriminatively learn the global priors. We validate our approach with a user study and compare against the state of the art, where we show significant improvements. Furthermore, we demonstrate our method extensively on many different types of images, including black-and-white photography from over a hundred years ago, and show realistic colorizations.

Single image 3D human pose estimation from noisy observations

Markerless 3D human pose detection from a single image is a severely underconstrained problem because different 3D poses can have similar image projections. In order to handle this ambiguity, current approaches rely on prior shape models that can only be correctly adjusted if 2D image features are accurately detected. Unfortunately, although current 2D part detector algorithms have shown promising results, they are not yet accurate enough to guarantee a complete disambiguation of the 3D inferred shape. In this paper, we introduce a novel approach for estimating 3D human pose even when observations are noisy. We propose a stochastic sampling strategy to propagate the noise from the image plane to the shape space. This provides a set of ambiguous 3D shapes, which are virtually undistinguishable from their image projections. Disambiguation is then achieved by imposing kinematic constraints that guarantee the resulting pose resembles a 3D human shape. We validate the method on a variety of situations in which state-of-the-art 2D detectors yield either inaccurate estimations or partly miss some of the body parts.

Globally and locally consistent image completion

We present a novel approach for image completion that results in images that are both locally and globally consistent. With a fully-convolutional neural network, we can complete images of arbitrary resolutions by filling-in missing regions of any shape. To train this image completion network to be consistent, we use global and local context discriminators that are trained to distinguish real images from completed ones. The global discriminator looks at the entire image to assess if it is coherent as a whole, while the local discriminator looks only at a small area centered at the completed region to ensure the local consistency of the generated patches. The image completion network is then trained to fool the both context discriminator networks, which requires it to generate images that are indistinguishable from real ones with regard to overall consistency as well as in details. We show that our approach can be used to complete a wide variety of scenes. Furthermore, in contrast with the patch-based approaches such as PatchMatch, our approach can generate fragments that do not appear elsewhere in the image, which allows us to naturally complete the images of objects with familiar and highly specific structures, such as faces.

A Joint Model for 2D and 3D Pose Estimation from a Single Image

We introduce a novel approach to automatically recover 3D human pose from a single image. Most previous work follows a pipelined approach: initially, a set of 2D features such as edges, joints or silhouettes are detected in the image, and then these observations are used to infer the 3D pose. Solving these two problems separately may lead to erroneous 3D poses when the feature detector has performed poorly. In this paper, we address this issue by jointly solving both the 2D detection and the 3D inference problems. For this purpose, we propose a Bayesian framework that integrates a generative model based on latent variables and discriminative 2D part detectors based on HOGs, and perform inference using evolutionary algorithms. Real experimentation demonstrates competitive results, and the ability of our methodology to provide accurate 2D and 3D pose estimations even when the 2D detectors are inaccurate.

Neuroaesthetics in fashion: Modeling the perception of fashionability

In this paper, we analyze the fashion of clothing of a large social website. Our goal is to learn and predict how fashionable a person looks on a photograph and suggest subtle improvements the user could make to improve her/his appeal. We propose a Conditional Random Field model that jointly reasons about several fashionability factors such as the type of outfit and garments the user is wearing, the type of the user, the photographs setting (e.g., the scenery behind the user), and the fashionability score. Importantly, our model is able to give rich feedback back to the user, conveying which garments or even scenery she/he should change in order to improve fashionability. We demonstrate that our joint approach significantly outperforms a variety of intelligent baselines. We additionally collected a novel heterogeneous dataset with 144,169 user posts containing diverse image, textual and meta information which can be exploited for our task. We also provide a detailed analysis of the data, showing different outfit trends and fashionability scores across the globe and across a span of 6 years.

Learning to simplify: fully convolutional networks for rough sketch cleanup

In this paper, we present a novel technique to simplify sketch drawings based on learning a series of convolution operators. In contrast to existing approaches that require vector images as input, we allow the more general and challenging input of rough raster sketches such as those obtained from scanning pencil sketches. We convert the rough sketch into a simplified version which is then amendable for vectorization. This is all done in a fully automatic way without user intervention. Our model consists of a fully convolutional neural network which, unlike most existing convolutional neural networks, is able to process images of any dimensions and aspect ratio as input, and outputs a simplified sketch which has the same dimensions as the input image. In order to teach our model to simplify, we present a new dataset of pairs of rough and simplified sketch drawings. By leveraging convolution operators in combination with efficient use of our proposed dataset, we are able to train our sketch simplification model. Our approach naturally overcomes the limitations of existing methods, e.g., vector images as input and long computation time; and we show that meaningful simplifications can be obtained for many different test cases. Finally, we validate our results with a user study in which we greatly outperform similar approaches and establish the state of the art in sketch simplification of raster images.

A high performance CRF model for clothes parsing

In this paper we tackle the problem of clothing parsing: Our goal is to segment and classify different garments a person is wearing. We frame the problem as the one of inference in a pose-aware Conditional Random Field (CRF) which exploits appearance, figure/ground segmentation, shape and location priors for each garment as well as similarities between segments, and symmetries between different human body parts. We demonstrate the effectiveness of our approach on the Fashionista dataset [1] and show that we can obtain a significant improvement over the state-of-the-art.

Fashion Style in 128 Floats: Joint Ranking and Classification Using Weak Data for Feature Extraction

We propose a novel approach for learning features from weakly-supervised data by joint ranking and classification. In order to exploit data with weak labels, we jointly train a feature extraction network with a ranking loss and a classification network with a cross-entropy loss. We obtain high-quality compact discriminative features with few parameters, learned on relatively small datasets without additional annotations. This enables us to tackle tasks with specialized images not very similar to the more generic ones in existing fully-supervised datasets. We show that the resulting features in combination with a linear classifier surpass the state-of-the-art on the Hipster Wars dataset despite using features only 0.3% of the size. Our proposed features significantly outperform those obtained from networks trained on ImageNet, despite being 32 times smaller (128 single-precision floats), trained on noisy and weakly-labeled data, and using only 1.5% of the number of parameters.

DaLI: Deformation and Light Invariant Descriptor

Recent advances in 3D shape analysis and recognition have shown that heat diffusion theory can be effectively used to describe local features of deforming and scaling surfaces. In this paper, we show how this description can be used to characterize 2D image patches, and introduce DaLI, a novel feature point descriptor with high resilience to non-rigid image transformations and illumination changes. In order to build the descriptor, 2D image patches are initially treated as 3D surfaces. Patches are then described in terms of a heat kernel signature, which captures both local and global information, and shows a high degree of invariance to non-linear image warps. In addition, by further applying a logarithmic sampling and a Fourier transform, invariance to photometric changes is achieved. Finally, the descriptor is compacted by mapping it onto a low dimensional subspace computed using Principal Component Analysis, allowing for an efficient matching. A thorough experimental validation demonstrates that DaLI is significantly more discriminative and robust to illuminations changes and image transformations than state of the art descriptors, even those specifically designed to describe non-rigid deformations.