Abstract: We present a vision-inertial system which combines two RGB-Depth devices together with a wearable inertial movement unit in order to detect activities of the daily living. From multi-view videos, we extract dense trajectories enriched with a histogram of normals description computed from the depth cue and bag them into multi-view codebooks. During the later classification step a multi-class support vector machine with a RBF- 2 kernel combines the descriptions at kernel level. In order to perform action detection from the videos, a sliding window approach is utilized. On the other hand, we extract accelerations, rotation angles, and jerk features from the inertial data collected by the wearable placed on the user’s dominant wrist. During gesture spotting, a dynamic time warping is applied and the aligning costs to a set of pre-selected gesture sub-classes are thresholded to determine possible detections. The outputs of the two modules are combined in a late-fusion fashion. The system is validated in a real-case scenario with elderly from an elder home. Learning-based fusion results improve the ones from the single modalities, demonstrating the success of such multimodal approach.

Abstract: The problem of removing illuminant variations to preserve the colours of objects (colour constancy) has already been solved by the human brain using mechanisms that rely largely on centre-surround computations of local contrast. In this paper we adopt some of these biological solutions described by long known physiological findings into a simple, fully automatic, functional model (termed Adaptive Surround Modulation or ASM). In ASM, the size of a visual neuron's receptive field (RF) as well as the relationship with its surround varies according to the local contrast within the stimulus, which in turn determines the nature of the centre-surround normalisation of cortical neurons higher up in the processing chain. We modelled colour constancy by means of two overlapping asymmetric Gaussian kernels whose sizes are adapted based on the contrast of the surround pixels, resembling the change of RF size. We simulated the contrast-dependent surround modulation by weighting the contribution of each Gaussian according to the centre-surround contrast. In the end, we obtained an estimation of the illuminant from the set of the most activated RFs' outputs. Our results on three single-illuminant and one multi-illuminant benchmark datasets show that ASM is highly competitive against the state-of-the-art and it even outperforms learning-based algorithms in one case. Moreover, the robustness of our model is more tangible if we consider that our results were obtained using the same parameters for all datasets, that is, mimicking how the human visual system operates. These results might provide an insight on how dynamical adaptation mechanisms contribute to make object's colours appear constant to us.

Abstract: Egocentric vision consists in acquiring images along the day from a first person point-of-view using wearable cameras. The automatic analysis of this information allows to discover daily patterns for improving the quality of life of the user. A natural topic that arises in egocentric vision is storytelling, that is, how to understand and tell the story relying behind the pictures.
In this paper, we tackle storytelling as an egocentric sequences description problem. We propose a novel methodology that exploits information from temporally neighboring events, matching precisely the nature of egocentric sequences. Furthermore, we present a new method for multimodal data fusion consisting on a multi-input attention recurrent network. We also release the EDUB-SegDesc dataset. This is the first dataset for egocentric image sequences description, consisting of 1,339 events with 3,991 descriptions, from 55 days acquired by 11 people. Finally, we prove that our proposal outperforms classical attentional encoder-decoder methods for video description.

Abstract: This paper presents a supervised subspace learning method called Kernel Generalized Discriminative Common Vectors (KGDCV), as a novel extension of the known Discriminative Common Vectors method with Kernels. Our method combines the advantages of kernel methods to model complex data and solve nonlinear
problems with moderate computational complexity, with the better generalization properties of generalized approaches for large dimensional data. These attractive combination makes KGDCV specially suited for feature extraction and classification in computer vision, image processing and pattern recognition applications. Two different approaches to this generalization are proposed, a first one based on the kernel trick (KT) and a second one based on the nonlinear projection trick (NPT) for even higher efficiency. Both methodologies
have been validated on four different image datasets containing faces, objects and handwritten digits, and compared against well known non-linear state-of-art methods. Results show better discriminant properties than other generalized approaches both linear or kernel. In addition, the KGDCV-NPT approach presents a considerable computational gain, without compromising the accuracy of the model.

Abstract: Target encoding plays a central role when learning Convolutional Neural Networks. In this realm, one-hot encoding is the most prevalent strategy due to its simplicity. However, this so widespread encoding schema assumes a flat label space, thus ignoring rich relationships existing among labels that can be exploited during training. In large-scale datasets, data does not span the full label space, but instead lies in a low-dimensional output manifold. Following this observation, we embed the targets into a low-dimensional space, drastically improving convergence speed while preserving accuracy. Our contribution is two fold: (i) We show that random projections of the label space are a valid tool to find such lower dimensional embeddings, boosting dramatically convergence rates at zero computational cost; and (ii) we propose a normalized eigenrepresentation of the class manifold that encodes the targets with minimal information loss, improving the accuracy of random projections encoding while enjoying the same convergence rates. Experiments on CIFAR-100, CUB200-2011, Imagenet, and MIT Places demonstrate that the proposed approach drastically improves convergence speed while reaching very competitive accuracy rates.

Abstract: This paper presents the development of an Augmented Reality mobile application which aims at sensibilizing young children to abstract concepts of music. Such concepts are, for instance, the musical notation or the idea of rhythm. Recent studies in Augmented Reality for education suggest that such technologies have multiple benefits for students, including younger ones. As mobile document image acquisition and processing gains maturity on mobile platforms, we explore how it is possible to build a markerless and real-time application to augment the physical documents with didactic animations and interactive virtual content. Given a standard image processing pipeline, we compare the performance of different local descriptors at two key stages of the process. Results suggest alternatives to the SIFT local descriptors, regarding result quality and computational efficiency, both for document model identification and perspective transform estimation. All experiments are performed on an original and public dataset we introduce here.

Abstract: This guest editorial introduces the twenty two papers accepted for this Special Issue on Articulated Motion and Deformable Objects (AMDO). They are grouped into four main categories within the field of AMDO: human motion analysis (action/gesture), human pose estimation, deformable shape segmentation, and face analysis. For each of the four topics, a survey of the recent developments in the field is presented. The accepted papers are briefly introduced in the context of this survey. They contribute novel methods, algorithms with improved performance as measured on benchmarking datasets, as well as two new datasets for hand action detection and human posture analysis. The special issue should be of high relevance to the reader interested in AMDO recognition and promote future research directions in the field.

Abstract: State-of-the-art approaches on hand pose estimation from depth images have reported promising results under quite controlled considerations. In this paper we propose a two-step pipeline for recovering the hand pose from a sequence of depth images. The pipeline has been designed to deal with images taken from any viewpoint and exhibiting a high degree of finger occlusion. In a first step we initialize the hand pose using a part-based model, fitting a set of hand components in the depth images. In a second step we consider temporal data and estimate the parameters of a trained bilinear model consisting of shape and trajectory bases. We evaluate our approach on a new created synthetic hand dataset along with NYU and MSRA real datasets. Results demonstrate that the proposed method outperforms the most recent pose recovering approaches, including those based on CNNs.