We present a method for prediction of a person's hairstyle from a single image. Despite growing use cases in user digitization and enrollment for virtual experiences, available methods are limited, particularly in the range of hairstyles they can capture. Human hair is extremely diverse and lacks any universally accepted description or categorization, making this a challenging task. Most current methods rely on parametric models of hair at a strand level. These approaches, while very promising, are not yet able to represent short, frizzy, coily hair and gathered hairstyles. We instead choose a classification approach which can represent the diversity of hairstyles required for a truly robust and inclusive system. Previous classification approaches have been restricted by poorly labeled data that lacks diversity, imposing constraints on the usefulness of any resulting enrollment system. We use only synthetic data to train our models. This allows for explicit control of diversity of hairstyle attributes, hair colors, facial appearance, poses, environments and other parameters. It also produces noise-free ground-truth labels. We introduce a novel hairstyle taxonomy developed in collaboration with a diverse group of domain experts which we use to balance our training data, supervise our model, and directly measure fairness. We annotate our synthetic training data and a real evaluation dataset using this taxonomy and release both to enable comparison of future hairstyle prediction approaches. We employ an architecture based on a pre-trained feature extraction network in order to improve generalization of our method to real data and predict taxonomy attributes as an auxiliary task to improve accuracy. Results show our method to be significantly more robust for challenging hairstyles than recent parametric approaches.

Deep neural networks and decision trees operate on largely separate paradigms; typically, the former performs representation learning with pre-specified architectures, while the latter is characterised by learning hierarchies over pre-specified features with data-driven architectures. We unite the two via adaptive neural trees (ANTs), a model that incorporates representation learning into edges, routing functions and leaf nodes of a decision tree, along with a backpropagation-based training algorithm that adaptively grows the architecture from primitive modules (e.g., convolutional layers). We demonstrate that, whilst achieving competitive performance on classification and regression datasets, ANTs benefit from (i) lightweight inference via conditional computation, (ii) hierarchical separation of features useful to the predictive task e.g. learning meaningful class associations, such as separating natural vs. man-made objects, and (iii) a mechanism to adapt the architecture to the size and complexity of the training dataset.

We present a novel cost function for semi-supervised learning of neural networks that encourages compact clustering of the latent space to facilitate separation. The key idea is to dynamically create a graph over embeddings of labeled and unlabeled samples of a training batch to capture underlying structure in feature space, and use label propagation to estimate its high and low density regions. We then devise a cost function based on Markov chains on the graph that regularizes the latent space to form a single compact cluster per class, while avoiding to disturb existing clusters during optimization. We evaluate our approach on three benchmarks and compare to state-of-the art with promising results. Our approach combines the benefits of graph-based regularization with efficient, inductive inference, does not require modifications to a network architecture, and can thus be easily applied to existing networks to enable an effective use of unlabeled data.

We propose a new method for creating computationally efficient and compact convolutional neural networks (CNNs) using a novel sparse connection structure that resembles a tree root. This allows a significant reduction in computational cost and number of parameters compared to state-of-the-art deep CNNs, without compromising accuracy, by exploiting the sparsity of inter-layer filter dependencies. We validate our approach by using it to train more efficient variants of state-of-the-art CNN architectures, evaluated on the CIFAR10 and ILSVRC datasets. Our results show similar or higher accuracy than the baseline architectures with much less computation, as measured by CPU and GPU timings. For example, for ResNet 50, our model has 40% fewer parameters, 45% fewer floating point operations, and is 31% (12%) faster on a CPU (GPU). For the deeper ResNet 200 our model has 48% fewer parameters and 27% fewer floating point operations, while maintaining state-of-the-art accuracy. For GoogLeNet, our model has 7% fewer parameters and is 21% (16%) faster on a CPU (GPU).

We investigate uncertainty quantification under a sparse Bayesian model of medical image registration. Bayesian modelling has proven powerful to automate the tuning of registration hyperparameters, such as the trade-off between the data and regularization functionals. Sparsity-inducing priors have recently been used to render the parametrization itself adaptive and data-driven. The sparse prior on transformation parameters effectively favors the use of coarse basis functions to capture the global trends in the visible motion while finer, highly localized bases are introduced only in the presence of coherent image information and motion. In earlier work, approximate inference under the sparse Bayesian model was tackled in an efficient Variational Bayes (VB) framework. In this paper we are interested in the theoretical and empirical quality of uncertainty estimates derived under this approximate scheme vs. under the exact model. We implement an (asymptotically) exact inference scheme based on reversible jump Markov Chain Monte Carlo (MCMC) sampling to characterize the posterior distribution of the transformation and compare the predictions of the VB and MCMC based methods. The true posterior distribution under the sparse Bayesian model is found to be meaningful: orders of magnitude for the estimated uncertainty are quantitatively reasonable, the uncertainty is higher in textureless regions and lower in the direction of strong intensity gradients.

We consider the task of predicting various traits of a person given an image of their face. We aim to estimate traits such as gender, ethnicity and age, as well as more subjective traits as the emotion a person expresses or whether they are humorous or attractive. Due to the recent surge of research on Deep Convolutional Neural Networks (CNNs), we begin by using a CNN architecture, and corroborate that CNNs are promising for facial attribute prediction. To further improve performance, we propose a novel approach that incorporates facial landmark information for input images as an additional channel, helping the CNN learn face-specific features so that the landmarks across various training images hold correspondence. We empirically analyze the performance of our proposed method, showing consistent improvement over the baselines across traits. We demonstrate our system on a sizeable Face Attributes Dataset (FAD), comprising of roughly 200,000 labels, for 10 most sought-after traits, for over 10,000 facial images.

Deep Convolutional Neural Networks (CNNs) have recently evinced immense success for various image recognition tasks [11, 27]. However, a question of paramount importance is somewhat unanswered in deep learning research - is the selected CNN optimal for the dataset in terms of accuracy and model size? In this paper, we intend to answer this question and introduce a novel strategy that alters the architecture of a given CNN for a specified dataset, to potentially enhance the original accuracy while possibly reducing the model size. We use two operations for architecture refinement, viz. stretching and symmetrical splitting. Stretching increases the number of hidden units (nodes) in a given CNN layer, while a symmetrical split of say K between two layers separates the input and output channels into K equal groups, and connects only the corresponding input-output channel groups. Our procedure starts with a pre-trained CNN for a given dataset, and optimally decides the stretch and split factors across the network to refine the architecture. We empirically demonstrate the necessity of the two operations. We evaluate our approach on two natural scenes attributes datasets, SUN Attributes [16] and CAMIT-NSAD [20], with architectures of GoogleNet and VGG-11, that are quite contrasting in their construction. We justify our choice of datasets, and show that they are interestingly distinct from each other, and together pose a challenge to our architectural refinement algorithm. Our results substantiate the usefulness of the proposed method.

Despite having high accuracy, neural nets have been shown to be susceptible to adversarial examples, where a small perturbation to an input can cause it to become mislabeled. We propose metrics for measuring the robustness of a neural net and devise a novel algorithm for approximating these metrics based on an encoding of robustness as a linear program. We show how our metrics can be used to evaluate the robustness of deep neural nets with experiments on the MNIST and CIFAR-10 datasets. Our algorithm generates more informative estimates of robustness metrics compared to estimates based on existing algorithms. Furthermore, we show how existing approaches to improving robustness “overfit” to adversarial examples generated using a specific algorithm. Finally, we show that our techniques can be used to additionally improve neural net robustness both according to the metrics that we propose, but also according to previously proposed metrics.

The use of depth-sensing computer vision to capture bodily movement is increasingly being exploited in healthcare. Yet, there are few descriptions of how real-world practices influence the design of such applications. To this end, we present the development and empirical evaluation of ASSESS MS, a system to support the clinical assessment of Multiple Sclerosis using Kinect. A key issue for developing machine-learning based systems is the need for standardized data on which statistical inferences can be made. We demonstrate that there are many aspects of clinical practice that are at odds with the need to capture standardized data for a computer vision system. We offer three design guidelines so address these: 1) Standardization is a multi-disciplinary issue and needs to be addressed early in the development process; 2) Tools that provide a view into what the camera “sees” can support the achievement of standardized data capture in real environments; 3) Tools to support standardized data capture should maintain the agency of human interaction. More broadly we show that when considering every day contexts, the traditional focus on measurement accuracy is only a small part of the effort needed to make a technology “work” in practice.

We present Deep Neural Decision Forests - a novel approach that unifies classification trees with the representation learning functionality known from deep convolutional networks, by training them in an end-to-end manner. To combine these two worlds, we introduce a stochastic and differentiable decision tree model, which steers the representation learning usually conducted in the initial layers of a (deep) convolutional network. Our model differs from conventional deep networks because a decision forest provides the final predictions and it differs from conventional decision forests since we propose a principled, joint and global optimization of split and leaf node parameters. We show experimental results on benchmark machine learning datasets like MNIST and ImageNet and find on-par or superior results when compared to state-of-the-art deep models. Most remarkably, we obtain Top5-Errors of only 7.84%/6.38% on ImageNet validation data when integrating our forests in a single-crop, single/seven model GoogLeNet architecture, respectively. Thus, even without any form of training data set augmentation we are improving on the 6.67% error obtained by the best GoogLeNet architecture (7 models, 144 crops).

In this paper we report the set-up and results of the Multimodal Brain Tumor Image Segmentation Benchmark (BRATS) organized in conjunction with the MICCAI 2012 and 2013 conferences. Twenty state-of-the-art tumor segmentation algorithms were applied to a set of 65 multi-contrast MR scans of low- and high-grade glioma patients-manually annotated by up to four raters-and to 65 comparable scans generated using tumor image simulation software. Quantitative evaluations revealed considerable disagreement between the human raters in segmenting various tumor sub-regions (Dice scores in the range 74%-85%), illustrating the difficulty of this task. We found that different algorithms worked best for different sub-regions (reaching performance comparable to human inter-rater variability), but that no single algorithm ranked in the top for all sub-regions simultaneously. Fusing several good algorithms using a hierarchical majority vote yielded segmentations that consistently ranked above all individual algorithms, indicating remaining opportunities for further methodological improvements. The BRATS image data and manual annotations continue to be publicly available through an online evaluation system as an ongoing benchmarking resource.

We present a new interactive and online approach to 3D scene understanding. Our system, SemanticPaint , allows users to simultaneously scan their environment whilst interactively segmenting the scene simply by reaching out and touching any desired object or surface. Our system continuously learns from these segmentations, and labels new unseen parts of the environment. Unlike offline systems where capture, labeling, and batch learning often take hours or even days to perform, our approach is fully online. This provides users with continuous live feedback of the recognition during capture, allowing to immediately correct errors in the segmentation and/or learning—a feature that has so far been unavailable to batch and offline methods. This leads to models that are tailored or personalized specifically to the user's environments and object classes of interest, opening up the potential for new applications in augmented reality, interior design, and human/robot navigation. It also provides the ability to capture substantial labeled 3D datasets for training large-scale visual recognition systems.

Recovery from tracking failure is essential in any simultaneous localization and tracking system. In this context, we explore an efficient keyframe-based relocalization method based on frame encoding using randomized ferns. The method enables automatic discovery of keyframes through online harvesting in tracking mode, and fast retrieval of pose candidates in the case when tracking is lost. Frame encoding is achieved by applying simple binary feature tests which are stored in the nodes of an ensemble of randomized ferns. The concatenation of small block codes generated by each fern yields a global compact representation of camera frames. Based on those representations we define the frame dissimilarity as the block-wise hamming distance (BlockHD). Dissimilarities between an incoming query frame and a large set of keyframes can be efficiently evaluated by simply traversing the nodes of the ferns and counting image co-occurrences in corresponding code tables. In tracking mode, those dissimilarities decide whether a frame/pose pair is considered as a novel keyframe. For tracking recovery, poses of the most similar keyframes are retrieved and used for reinitialization of the tracking algorithm. The integration of our relocalization method into a hand-held KinectFusion system allows seamless continuation of mapping even when tracking is frequently lost.

We propose 'filter forests' (FF), an efficient new discriminative approach for predicting continuous variables given a signal and its context. FF can be used for general signal restoration tasks that can be tackled via convolutional filtering, where it attempts to learn the optimal filtering kernels to be applied to each data point. The model can learn both the size of the kernel and its values, conditioned on the observation and its spatial or temporal context. We show that FF compares favorably to both Markov random field based and recently proposed regression forest based approaches for labeling problems in terms of efficiency and accuracy. In particular, we demonstrate how FF can be used to learn optimal denoising filters for natural images as well as for other tasks such as depth image refinement, and 1D signal magnitude estimation. Numerous experiments and quantitative comparisons show that FFs achieve accuracy at par or superior to recent state of the art techniques, while being several orders of magnitude faster.

We propose a Sparse Bayesian framework for non-rigid registration. Our principled approach is flexible, in that it efficiently finds an optimal, sparse model to represent deformations among any preset, widely overcomplete range of basis functions. It addresses open challenges in state-of-the-art registration, such as the automatic joint estimate of model parameters ( e.g. noise and regularization levels). We demonstrate the feasibility and performance of our approach on cine MR, tagged MR and 3D US cardiac images, and show state-of-the-art results on benchmark datasets evaluating accuracy of motion and strain. These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

We present a machine learning technique for estimating absolute, per-pixel depth using any conventional monocular 2D camera, with minor hardware modifications. Our approach targets close-range human capture and interaction where dense 3D estimation of hands and faces is desired. We use hybrid classification-regression forests to learn how to map from near infrared intensity images to absolute , metric depth in real-time. We demonstrate a variety of human-computer interaction and capture scenarios. Experiments show an accuracy that outperforms a conventional light fall-off baseline, and is comparable to high-quality consumer depth cameras, but with a dramatically reduced cost, power consumption, and form-factor.

Conventional decision forest based methods for image labelling tasks like object segmentation make predictions for each variable (pixel) independently [3, 5, 8]. This prevents them from enforcing dependencies between variables and translates into locally inconsistent pixel labellings. Random field models, instead, encourage spatial consistency of labels at increased computational expense. This paper presents a new and efficient forest based model that achieves spatially consistent semantic image segmentation by encoding variable dependencies directly in the feature space the forests operate on. Such correlations are captured via new long-range, soft connectivity features, computed via generalized geodesic distance transforms. Our model can be thought of as a generalization of the successful Semantic Texton Forest, Auto-Context, and Entangled Forest models. A second contribution is to show the connection between the typical Conditional Random Field (CRF) energy and the forest training objective. This analysis yields a new objective for training decision forests that encourages more accurate structured prediction. Our GeoF model is validated quantitatively on the task of semantic image segmentation, on four challenging and very diverse image datasets. GeoF outperforms both state of-the-art forest models and the conventional pair wise CRF.

We address the problem of inferring the pose of an RGB-D camera relative to a known 3D scene, given only a single acquired image. Our approach employs a regression forest that is capable of inferring an estimate of each pixel's correspondence to 3D points in the scene's world coordinate frame. The forest uses only simple depth and RGB pixel comparison features, and does not require the computation of feature descriptors. The forest is trained to be capable of predicting correspondences at any pixel, so no interest point detectors are required. The camera pose is inferred using a robust optimization scheme. This starts with an initial set of hypothesized camera poses, constructed by applying the forest at a small fraction of image pixels. Preemptive RANSAC then iterates sampling more pixels at which to evaluate the forest, counting inliers, and refining the hypothesized poses. We evaluate on several varied scenes captured with an RGB-D camera and observe that the proposed technique achieves highly accurate relocalization and substantially out-performs two state of the art baselines.

Image segmentation is the process of partitioning an image into segments or subsets of pixels for purposes of further analysis, such as separating the interesting objects in the foreground from the un-interesting objects in the background. In many image processing applications, the process requires a sequence of computational steps on a per pixel basis, thereby binding the performance to the size and resolution of the image. As applications require greater resolution and larger images the computational resources of this step can quickly exceed those of available CPUs, especially in the power and thermal constrained areas of consumer electronics and mobile. In this work, we use a hardware tree-based classifier to solve the image segmentation problem. The application is background removal (BGR) from depth-maps obtained from the Microsoft Kinect sensor. After the image is segmented, subsequent steps then classify the objects in the scene. The approach is flexible: to address different application domains we only need to change the trees used by the classifiers. We describe two distinct approaches and evaluate their performance using the commercial-grade testing environment used for the Microsoft Xbox gaming console.

We introduce an efficient camera relocalization approach which can be easily integrated into real-time 3D reconstruction methods, such as KinectFusion. Our approach makes use of compact encoding of whole image frames which enables both online harvesting of keyframes in tracking mode, and fast retrieval of pose proposals when tracking is lost. The encoding scheme is based on randomized ferns and simple binary feature tests. Each fern generates a small block code, and the concatenation of codes yields a compact representation of each camera frame. Based on those representations we introduce an efficient frame dissimilarity measure which is defined via the block-wise hamming distance (BlockHD). We illustrate how BlockHDs between a query frame and a large set of keyframes can be simultaneously evaluated by traversing the nodes of the ferns and counting image co-occurrences in corresponding code tables. In tracking mode, this mechanism allows us to consider every frame/pose pair as a potential keyframe. A new keyframe is added only if it is sufficiently dissimilar from all previously stored keyframes. For tracking recovery, camera poses are retrieved that correspond to the keyframes with smallest BlockHDs. The pose proposals are then used to reinitialize the tracking algorithm. Harvesting of keyframes and pose retrieval are computationally efficient with only small impact on the run-time performance of the 3D reconstruction. Integrating our relocalization method into KinectFusion allows seamless continuation of mapping even when tracking is frequently lost. Additionally, we demonstrate how marker-free augmented reality, in particular, can benefit from this integration by enabling a smoother and continuous AR experience.

Randomized decision trees and forests have a rich history in machine learning and have seen considerable success in application, perhaps particularly so for computer vision. However, they face a fundamental limitation: given enough data, the number of nodes in decision trees will grow exponentially with depth. For certain applications, for example on mobile or embedded processors, memory is a limited resource, and so the exponential growth of trees limits their depth, and thus their potential accuracy. This paper proposes decision jungles, revisiting the idea of ensembles of rooted decision directed acyclic graphs (DAGs), and shows these to be compact and powerful discriminative models for classification. Unlike conventional decision trees that only allow one path to every node, a DAG in a decision jungle allows multiple paths from the root to each leaf. We present and compare two new node merging algorithms that jointly optimize both the features and the structure of the DAGs efficiently. During training, node splitting and node merging are driven by the minimization of exactly the same objective function, here the weighted sum of entropies at the leaves. Results on varied datasets show that, compared to decision forests and several other baselines, decision jungles require dramatically less memory while considerably improving generalization.

This paper presents a new distance for measuring shape dissimilarity between objects. Recent publications introduced the use of eigenvalues of the Laplace operator as compact shape descriptors. Here, we revisit the eigenvalues to define a proper distance, called Weighted Spectral Distance (WESD), for quantifying shape dissimilarity. The definition of WESD is derived through analyzing the heat trace. This analysis provides the proposed distance with an intuitive meaning and mathematically links it to the intrinsic geometry of objects. We analyze the resulting distance definition, present and prove its important theoretical properties. Some of these properties include: 1) WESD is defined over the entire sequence of eigenvalues yet it is guaranteed to converge, 2) it is a pseudometric, 3) it is accurately approximated with a finite number of eigenvalues, and 4) it can be mapped to the [0,1) interval. Last, experiments conducted on synthetic and real objects are presented. These experiments highlight the practical benefits of WESD for applications in vision and medical image analysis.

We describe two new approaches to human pose estimation. Both can quickly and accurately predict the 3D positions of body joints from a single depth image without using any temporal information. The key to both approaches is the use of a large, realistic, and highly varied synthetic set of training images. This allows us to learn models that are largely invariant to factors such as pose, body shape, field-of-view cropping, and clothing. Our first approach employs an intermediate body parts representation, designed so that an accurate per-pixel classification of the parts will localize the joints of the body. The second approach instead directly regresses the positions of body joints. By using simple depth pixel comparison features and parallelizable decision forests, both approaches can run super-real time on consumer hardware. Our evaluation investigates many aspects of our methods, and compares the approaches to each other and to the state of the art. Results on silhouettes suggest broader applicability to other imaging modalities.

Methods that leverage neighbourhood structures in high-dimensional image spaces have recently attracted attention. These approaches extract information from a new image using its "neighbours" in the image space equipped with an application-specific distance. Finding the neighbourhood of a given image is challenging due to large dataset sizes and costly distance evaluations. Furthermore, automatic neighbourhood search for a new image is currently not possible when the distance is based on ground truth annotations. In this article we present a general and efficient solution to these problems. "neighbourhood approximation forests" (NAF) is a supervised learning algorithm that approximates the neighbourhood structure resulting from an arbitrary distance. As NAF uses only image intensities to infer neighbours it can also be applied to distances based on ground truth annotations. We demonstrate NAF in two scenarios: (i) choosing neighbours with respect to a deformation-based distance, and (ii) age prediction from brain MRI. The experiments show NAF's approximation quality, computational advantages and use in different contexts.

In this paper we introduce Context-Sensitive Decision Forests - A new perspective to exploit contextual information in the popular decision forest framework for the object detection problem. They are tree-structured classifiers with the ability to access intermediate prediction (here: classification and regression) information during training and inference time. This intermediate prediction is available to each sample, which allows us to develop context-based decision criteria, used for refining the prediction process. In addition, we introduce a novel split criterion which in combination with a priority based way of constructing the trees, allows more accurate regression mode selection and hence improves the current context information. In our experiments, we demonstrate improved results for the task of pedestrian detection on the challenging TUD data set when compared to state-of-the-art methods.

Segmentation-based scores play an important role in the evaluation of computational tools in medical image analysis. These scores evaluate the quality of various tasks, such as image registration and segmentation, by measuring the similarity between two binary label maps. Commonly these measurements blend two aspects of the similarity: pose misalignments and shape discrepancies. Not being able to distinguish between these two aspects, these scores often yield similar results to a widely varying range of different segmentation pairs. Consequently, the comparisons and analysis achieved by interpreting these scores become questionable. In this paper, we address this problem by exploring a new segmentation-based score, called normalized Weighted Spectral Distance (nWSD), that measures only shape discrepancies using the spectrum of the Laplace operator. Through experiments on synthetic and real data we demonstrate that nWSD provides additional information for evaluating differences between segmentations, which is not captured by other commonly used scores. Our results demonstrate that when jointly used with other scores, such as Dice's similarity coefficient, the additional information provided by nWSD allows richer, more discriminative evaluations. We show for the task of registration that through this addition we can distinguish different types of registration errors. This allows us to identify the source of errors and discriminate registration results which so far had to be treated as being of similar quality in previous evaluation studies.

The growth of image-guided procedures in surgical settings has led to an increased need to interact with digital images under sterile conditions. Traditional touch-based interaction techniques present challenges for managing asepsis in these environments leading to suggestions that new touchless interaction techniques may provide a compelling set of alternatives. In this paper we explore the potential for touchless interaction in image-guided Interventional Radiology (IR) through an ethnographic study. The findings highlight how the distribution of labour and spatial practices of this work are organised with respect to concerns about asepsis and radiation exposure, the physical and cognitive demands of artefact manipulation, patient management, and the construction of "professional vision". We discuss the implications of these key features of the work for touchless interaction technologies within IR and suggest that such issues will be of central importance in considering new input techniques in other medical settings.

We present a new approach to general-activity human pose estimation from depth images, building on Hough forests. We extend existing techniques in several ways: real time prediction of multiple 3D joints, explicit learning of voting weights, vote compression to allow larger training sets, and a comparison of several decision-tree training objectives. Key aspects of our work include: regression directly from the raw depth image, without the use of an arbitrary intermediate representation; applicability to general motions (not constrained to particular activities) and the ability to localize occluded as well as visible body joints. Experimental results demonstrate that our method produces state of the art results on several data sets including the challenging MSRC-5000 pose estimation test set, at a speed of about 200 frames per second. Results on silhouettes suggest broader applicability to other imaging modalities.

The objective of this work is to automatically generate a large number of images for a specified object class. A multimodal approach employing both text, metadata, and visual features is used to gather many high-quality images from the Web. Candidate images are obtained by a text-based Web search querying on the object identifier (e.g., the word penguin). The Webpages and the images they contain are downloaded. The task is then to remove irrelevant images and rerank the remainder. First, the images are reranked based on the text surrounding the image and metadata features. A number of methods are compared for this reranking. Second, the top-ranked images are used as (noisy) training data and an SVM visual classifier is learned to improve the ranking further. We investigate the sensitivity of the cross-validation procedure to this noisy training data. The principal novelty of the overall method is in combining text/metadata and visual features in order to achieve a completely automatic ranking of the images. Examples are given for a selection of animals, vehicles, and other classes, totaling 18 classes. The results are assessed by precision/recall curves on ground-truth annotated data and by comparison to previous approaches, including those of Berg and Forsyth and Fergus et al.

This paper presents an automatic segmentation algorithm for video frames captured by a (monocular) webcam that closely approximates depth segmentation from a stereo camera. The frames are segmented into foreground and background layers that comprise a subject (participant) and other objects and individuals. The algorithm produces correct segmentations even in the presence of large background motion with a nearly stationary foreground. This research makes three key contributions: First, we introduce a novel motion representation, referred to as "motons," inspired by research in object recognition. Second, we propose estimating the segmentation likelihood from the spatial context of motion. The estimation is efficiently learned by random forests. Third, we introduce a general taxonomy of tree-based classifiers that facilitates both theoretical and experimental comparisons of several known classification algorithms and generates new ones. In our bilayer segmentation algorithm, diverse visual cues such as motion, motion context, color, contrast, and spatial priors are fused by means of a conditional random field (CRF) model. Segmentation is then achieved by binary min-cut. Experiments on many sequences of our videochat application demonstrate that our algorithm, which requires no initialization, is effective in a variety of scenes, and the segmentation results are comparable to those obtained by stereo systems.

In this paper we introduce a new shape constraint for interactive image segmentation. It is an extension of Veksler's star-convexity prior, in two ways: from a single star to multiple stars and from Euclidean rays to Geodesic paths. Global minima of the energy function are obtained subject to these new constraints. We also introduce Geodesic Forests, which exploit the structure of shortest paths in implementing the extended constraints. The star-convexity prior is used here in an interactive setting and this is demonstrated in a practical system. The system is evaluated by means of a “robot user” to measure the amount of interaction required in a precise way. We also introduce a new and harder dataset which augments the existing Grabcut dataset with images and ground truth taken from the PASCAL VOC segmentation challenge.

This article presents a new, unified technique to perform general edge-sensitive editing operations on n-dimensional images and videos efficiently. The first contribution of the article is the introduction of a Generalized Geodesic Distance Transform (GGDT), based on soft masks. This provides a unified framework to address several edge-aware editing operations. Diverse tasks such as denoising and nonphotorealistic rendering are all dealt with fundamentally the same, fast algorithm. Second, a new Geodesic Symmetric Filter (GSF) is presented which imposes contrast-sensitive spatial smoothness into segmentation and segmentation-based editing tasks (cutout, object highlighting, colorization, panorama stitching). The effect of the filter is controlled by two intuitive, geometric parameters. In contrast to existing techniques, the GSF filter is applied to real-valued pixel likelihoods (soft masks), thanks to GGDTs and it can be used for both interactive and automatic editing. Complex object topologies are dealt with effortlessly. Finally, the parallelism of GGDTs enables us to exploit modern multicore CPU architectures as well as powerful new GPUs, thus providing great flexibility of implementation and deployment. Our technique operates on both images and videos, and generalizes naturally to n-dimensional data. The proposed algorithm is validated via quantitative and qualitative comparisons with existing, state-of-the-art approaches. Numerous results on a variety of image and video editing tasks further demonstrate the effectiveness of our method.

This paper presents a novel method for location recognition, which exploits an epitomic representation to achieve both high efficiency and good generalization. A generative model based on epitomic image analysis captures the appearance and geometric structure of an environment while allowing for variations due to motion, occlusions, and non-Lambertian effects. The ability to model translation and scale invariance together with the fusion of diverse visual features yields enhanced generalization with economical training. Experiments on both existing and new labeled image databases result in recognition accuracy superior to state of the art with real-time computational performance.

This work investigates the use of Random Forests for class based pixel-wise segmentation of images. The contribution of this paper is three-fold. First, we show that apparently quite dissimilar classifiers (such as nearest neighbour matching to texton class histograms) can be mapped onto a Random Forest architecture. Second, based on this insight, we show that the performance of such classifiers can be improved by incorporating the spatial context and discriminative learning that arises naturally in the Random Forest framework. Finally, we show that the ability of Random Forests to combine multiple features leads to a further increase in performance when textons, colour, filterbanks, and HOG features are used simultaneously. The benefit of the multi-feature classifier is demonstrated with extensive experimentation on existing labelled image datasets. The method equals or exceeds the state of the art on these datasets. 1

This paper presents a novel method for location recognition, which exploits an epitomic representation to achieve both high efficiency and good generalization. A generative model based on epitomic image analysis captures the appearance and geometric structure of an environment while allowing for variations due to motion, occlusions and non-Lambertian effects. The ability to model translation and scale invariance together with the fusion of diverse visual features yield enhanced generalization with economical training. Experiments on both existing and new labelled image databases result in recognition accuracy superior to state of the art with real-time computational performance.

A new method for localising and recognising hand poses and objects in real-time is presented. This problem is important in vision-driven applications where it is natural for a user to combine hand gestures and real objects when interacting with a machine. Examples include using a real eraser to remove words from a document displayed on an electronic surface. In this paper the task of simultaneously recognising object classes, hand gestures and detecting touch events is cast as a single classification problem. A random forest algorithm is employed which adaptively selects and combines a minimal set of appearance, shape and stereo features to achieve maximum class discrimination for a given image. This minimal set leads to both efficiency at run time and good generalisation. Unlike previous stereo works which explicitly construct disparity maps, here the stereo matching costs are used directly as visual cue and only computed on-demand, i.e. only for pixels where they are necessary for recognition. This leads to improved efficiency. The proposed method is assessed on a database of a variety of objects and hand poses selected for interacting on a flat surface in an office environment.

This paper presents an algorithm for the automatic segmentation of monocular videos into foreground and background layers. Correct segmentations are produced even in the presence of large background motion with nearly stationary foreground. There are three key contributions. The first is the introduction of a novel motion representation, "motons", inspired by research in object recognition. Second, we propose learning the segmentation likelihood from the spatial context of motion. The learning is efficiently performed by Random Forests. The third contribution is a general taxonomy of tree-based classifiers, which facilitates theoretical and experimental comparisons of several known classification algorithms, as well as spawning new ones. Diverse visual cues such as motion, motion context, colour, contrast and spatial priors are fused together by means of a conditional random field (CRF) model. Segmentation is then achieved by binary min-cut. Our algorithm requires no initialization. Experiments on many video-chat type sequences demonstrate the effectiveness of our algorithm in a variety of scenes. The segmentation results are comparable to those obtained by stereo systems.

Online camera recalibration is necessary for long-term deployment of computer vision systems. Existing algorithms assume that the source of recalibration information is a set of features in a general 3D scene; and that enough features are observed that the calibration problem is well-constrained. However; these assumptions are frequently invalid outside the laboratory. Real-world scenes often lack texture, contain repeated texture, or are mostly planar, making calibration difficult or impossible. In this paper we consider the calibration of families of stereo cameras, where each camera is assumed to have parameters drawn from a common but unknown prior distribution. We show how estimation of this prior using a small-number of offline-calibrated cameras (e.g. from the same production line) allows online calibration of additional cameras using a small number of point correspondences; and that using the estimated prior significantly increases the accuracy and robustness of stereo camera calibration.

The objective of this work is to automatically generate a large number of images for a specified object class (for example, penguin). A multi-modal approach employing both text, meta data and visual features is used to gather many, high-quality images from the Web. Candidate images are obtained by a text based Web search querying on the object identifier (the word penguin). The Web pages and the images they contain are downloaded. The task is then to remove irrelevant images and re-rank the remainder. First, the images are re-ranked using a Bayes posterior estimator trained on the text surrounding the image and meta data features (such as the image alternative tag, image title tag, and image filename). No visual information is used at this stage. Second, the top-ranked images are used as (noisy) training data and a SVM visual classifier is learnt to improve the ranking further. The principal novelty is in combining text/meta-data and visual features in order to achieve a completely automatic ranking of the images. Examples are given for a selection of animals (e.g. camels, sharks, penguins), vehicles (cars, airplanes, bikes) and other classes (guitar, wristwatch), totalling 18 classes. The results are assessed by precision/recall curves on ground truth annotated data and by comparison to previous approaches including those of Berg et al. (on an additional six classes) and Fergus et al.

We present a system for inserting new objects into existing photographs by querying a vast image-based object library, pre-computed using a publicly available Internet object database. The central goal is to shield the user from all of the arduous tasks typically involved in image compositing. The user is only asked to do two simple things: 1) pick a 3D location in the scene to place a new object; 2) select an object to insert using a hierarchical menu. We pose the problem of object insertion as a data-driven, 3D-based, context-sensitive object retrieval task. Instead of trying to manipulate the object to change its orientation, color distribution, etc. to fit the new image, we simply retrieve an object of a specified class that has all the required properties (camera pose, lighting, resolution, etc) from our large object library. We present new automatic algorithms for improving object segmentation and blending, estimating true 3D object size and orientation, and estimating scene lighting conditions. We also present an intuitive user interface that makes object insertion fast and simple even for the artistically challenged.

This paper presents an algorithm capable of real-time separation of foreground from background in monocular video sequences. Automatic segmentation of layers from colour/contrast or from motion alone is known to be error-prone. Here motion, colour and contrast cues are probabilistically fused together with spatial and temporal priors to infer layers accurately and efficiently. Central to our algorithm is the fact that pixel velocities are not needed, thus removing the need for optical flow estimation, with its tendency to error and computational expense. Instead, an efficient motion vs nonmotion classifier is trained to operate directly and jointly on intensity-change and contrast. Its output is then fused with colour information. The prior on segmentation is represented by a second order, temporal, Hidden Markov Model, together with a spatial MRF favouring coherence except where contrast is high. Finally, accurate layer segmentation and explicit occlusion detection are efficiently achieved by binary graph cut. The segmentation accuracy of the proposed algorithm is quantitatively evaluated with respect to existing groundtruth data and found to be comparable to the accuracy of a state of the art stereo segmentation algorithm. Foreground/ background segmentation is demonstrated in the application of live background substitution and shown to generate convincingly good quality composite video.

This paper presents a new model of object classes which incorporates appearance and shape information jointly. Modeling objects appearance by distributions of visual words has recently proven successful. Here appearancebased models are augmented by capturing the spatial arrangement of visual words. Compact spatial modeling without loss of discrimination is achieved through the introduction of adaptive vector quantized correlograms, which we call correlatons. Efficiency is further improved by means of integral images. The robustness of our new models to geometric transformations, severe occlusions and missing information is also demonstrated. The accuracy of discrimination of the proposed models is assessed with respect to existing databases with large numbers of object classes viewed under general conditions, and shown to outperform appearance-only models.

This paper describes models and algorithms for the real-time segmentation of foreground from background layers in stereo video sequences. Automatic separation of layers from color/contrast or from stereo alone is known to be error-prone. Here, color, contrast, and stereo matching information are fused to infer layers accurately and efficiently. The first algorithm, Layered Dynamic Programming (LDP), solves stereo in an extended six-state space that represents both foreground/background layers and occluded regions. The stereo-match likelihood is then fused with a contrast-sensitive color model that is learned on-the-fly and stereo disparities are obtained by dynamic programming. The second algorithm, Layered Graph Cut (LGC), does not directly solve stereo. Instead, the stereo match likelihood is marginalized over disparities to evaluate foreground and background hypotheses and then fused with a contrast-sensitive color model like the one used in LDP. Segmentation is solved efficiently by ternary graph cut. Both algorithms are evaluated with respect to ground truth data and found to have similar performance, substantially better than either stereo or color/ contrast alone. However, their characteristics with respect to computational efficiency are rather different. The algorithms are demonstrated in the application of background substitution and shown to give good quality composite video output.

This paper describes two algorithms capable of real-time segmentation of foreground from background layers in stereo video sequences. Automatic separation of layers from colour/contrast or from stereo alone is known to be error-prone. Here, colour, contrast and stereo matching information are fused to infer layers accurately and efficiently. The first algorithm, layered dynamic programming (LDP), solves stereo in an extended 6-state space that represents both foreground/background layers and occluded regions. The stereo-match likelihood is then fused with a contrast-sensitive colour model that is learned on the fly, and stereo disparities are obtained by dynamic programming. The second algorithm, layered graph cut (LGC), does not directly solve stereo. Instead the stereo match likelihood is marginalised over foreground and background hypotheses, and fused with a contrast-sensitive colour model like the one used in LDP. Segmentation is solved efficiently by ternary graph cut. Both algorithms are evaluated with respect to ground truth data and found to have similar p performance, substantially better than stereo or colour/contrast alone. However, their characteristics with respect to computational efficiency are rather different. The algorithms are demonstrated in the application of background substitution and shown to give good quality composite video output.

This paper demonstrates the high quality, real-time segmentation techniques. We achieve real-time segmentation of foreground from background layers in stereo video sequences. Automatic separation of layers from colour/contrast or from stereo alone is known to be error-prone. Here, colour, contrast and stereo matching information are fused to infer layers accurately and efficiently. The first algorithm, layered dynamic programming (LDP), solves stereo in an extended 6-state space that represents both foreground/background layers and occluded regions. The stereo-match likelihood is then fused with a contrast-sensitive colour model that is learned on the fly, and stereo disparities are obtained by dynamic programming. The second algorithm, layered graph cut (LGC), does not directly solve stereo. Instead the stereo match likelihood is marginalised over foreground and background hypotheses, and fused with a contrast-sensitive colour model like the one used in LDP. Segmentation is solved efficiently by ternary graph cut. Both algorithms are evaluated with respect to ground truth data and found to have similar performance, substantially better than stereo or colour/contrast alone. However, their characteristics with respect to computational efficiency are rather different. The algorithms are demonstrated in the application of background substitution and shown to give good quality composite video output.

This paper presents a new algorithm for the automatic recognition of object classes from images (categorization). Compact and yet discriminative appearance-based object class models are automatically learned from a set of training images. The method is simple and extremely fast, making it suitable for many applications such as semantic image retrieval, Web search, and interactive image editing. It classifies a region according to the proportions of different visual words (clusters in feature space). The specific visual words and the typical proportions in each object are learned from a segmented training set. The main contribution of this paper is twofold: i) an optimally compact visual dictionary is learned by pair-wise merging of visual words from an initially large dictionary. The final visual words are described by GMMs. ii) A novel statistical measure of discrimination is proposed which is optimized by each merge operation. High classification accuracy is demonstrated for nine object classes on photographs of real objects viewed under general lighting conditions, poses and viewpoints. The set of test images used for validation comprise: i) photographs acquired by us, ii) images from the Web and iii) images from the recently released Pascal dataset. The proposed algorithm performs well on both texture-rich objects (e.g. grass, sky, trees) and structure-rich ones (e.g. cars, bikes, planes)

A new algorithm is proposed for removing large objects from digital images. The challenge is to fill in the hole that is left behind in a visually plausible way. In the past, this problem has been addressed by two classes of algorithms: 1) "texture synthesis" algorithms for generating large image regions from sample textures and 2) "inpainting" techniques for filling in small image gaps. The former has been demonstrated for "textures"--repeating two-dimensional patterns with some stochasticity; the latter focus on linear "structures" which can be thought of as one-dimensional patterns, such as lines and object contours. This paper presents a novel and efficient algorithm that combines the advantages of these two approaches. We first note that exemplar-based texture synthesis contains the essential process required to replicate both texture and structure; the success of structure propagation, however, is highly dependent on the order in which the filling proceeds. We propose a best-first algorithm in which the confidence in the synthesized pixel values is propagated in a manner similar to the propagation of information in inpainting. The actual color values are computed using exemplar-based synthesis. In this paper, the simultaneous propagation of texture and structure information is achieved by a single, efficient algorithm. Computational efficiency is achieved by a block-based sampling process. A number of examples on real and synthetic images demonstrate the effectiveness of our algorithm in removing large occluding objects, as well as thin scratches. Robustness with respect to the shape of the manually selected target region is also demonstrated. Our results compare favorably to those obtained by existing techniques.

A new algorithm is proposed for removing large objects from digital images. The challenge is to fill in the hole that is left behind in a visually plausible way. In the past, this problem has been addressed by two classes of algorithms: (i) "texture synthesis" algorithms for generating large image regions from sample textures, and (ii) "inpainting" techniques for filling in small image gaps. The former work well for "textures" - repeating two dimensional patterns with some stochasticity; the latter focus on linear "structures" which can be thought of as one dimensional patterns, such as lines and object contours. This paper presents a novel and efficient algorithm that combines the advantages of these two approaches. We first note that exemplar-based texture synthesis contains the essential process required to replicate both texture and structure; the success of structure propagation, however, is highly dependent on the order in which the filling proceeds. We propose a best-first algorithm in which the confidence in the synthesized pixel values is propagated in a manner similar to the propagation of information in inpainting. The actual color values are computed using exemplar-based synthesis. Computational efficiency is achieved by a block-based sampling process. A number of examples on real and synthetic images demonstrate the effectiveness of our algorithm in removing large occluding objects as well as thin scratches. Robustness with respect to the shape of the manually selected target region is also demonstrated. Our results compare favorably to those obtained by existing techniques.

A new algorithm is proposed for novel view generation in one-to-one teleconferencing applications. Given the video streams acquired by two cameras placed on either side of a computer monitor, the proposed algorithm synthesizes images from a virtual camera in arbitrary position (typically located within the monitor) to facilitate eye contact. Our technique is based on an improved, dynamic-programming, stereo algorithm for efficient novel-view generation. The two main contributions are: i) a new type of three-plane graph for dense-stereo dynamic-programming, that encourages correct occlusion labeling; ii) a compact geometric derivation for novel-view synthesis by direct projection of the minimum-cost surface. Furthermore, we present a novel algorithm for the temporal maintenance of a background model to enhance the rendering of occlusions and reduce temporal artefacts (flicker); and a cost aggregation algorithm that acts directly on our three-dimensional matching cost space. Examples are given that demonstrate the robustness of the new algorithm to spatial and temporal artefacts for long stereo video streams. These include demonstrations of synthesis of cyclopean views of extended conversational sequences. We further demonstrate synthesis from a freely translating virtual camera.

The objective of this paper is to estimate the orientation of a scene plane from an uncal-ibrated perspective image under the assumption that the scene is coated with a homoge-neous (but unknown) texture. We make the following novel contributions: first, we show that the problem is equivalent to estimating the vanishing line of the plane; second, we show that estimating the two degrees of freedom of this line can be decomposed into two searches each for one parameter; third, we give an algorithm for this estimation which is applicable to both regular and irregular textures. The algorithms do not require that texels are identified explicitly. But once the plane vanishing line has been obtained, then texels locations can be determined, and the geometry of the scene plane computed up to an affine transformation. We give examples of these computations on real images. 1

We describe how 3D affine measurements may be computed from a single perspective view of a scene given only minimal geometric information determined from the image. This minimal information is typically the vanishing line of a reference plane and a vanishing point for a direction not parallel to the plane. It is shown that affine scene structure may then be determined from the image, without knowledge of the camera's internal calibration (e.g. focal length), nor of the explicit relation between camera and world (pose). In particular we show how to: compute the distance between planes parallel to the reference plane (up to a common scale factor); compute area and length ratios on any plane parallel to the reference plane; determine the camera's (viewer's) location. Simple geometric derivations are given for these results. We also develop an algebraic representation which unifies the three types of measurement and, amongst other advantages, permits a first order error propagation analysis to be performed, associating an uncertainty with each measurement. We demonstrate the technique for a variety of applications, including height measurements in forensic images and 3D graphical modelling from single images.

We present methods for creating 3D graphical models of scenes from a limited numbers of images, i.e. one or two, in situations where no scene co‐ordinate measurements are available. The methods employ constraints available from geometric relationships that are common in architectural scenes – such as parallelism and orthogonality – together with constraints available from the camera. In particular, by using the circular points of a plane simple, linear algorithms are given for computing plane rectification, plane orientation and camera calibration from a single image. Examples of image based 3D modelling are given for both single images and image pairs.