Abstract: Interest in automatic action and gesture recognition has grown considerably in the last few years. This is due in part to the large number of application domains for this type of technology. As in many other computer vision areas, deep learning based methods have quickly become a reference methodology for obtaining state-of-the-art performance in both tasks. This chapter is a survey of current deep learning based methodologies for action and gesture recognition in sequences of images. The survey reviews both fundamental and cutting edge methodologies reported in the last few years. We introduce a taxonomy that summarizes important aspects of deep learning for approaching both tasks. Details of the proposed architectures, fusion strategies, main datasets, and competitions are reviewed. Also, we summarize and discuss the main works proposed so far with particular interest on how they treat the temporal dimension of data, their highlighting features, and opportunities and challenges for future research. To the best of our knowledge this is the first survey in the topic. We foresee this survey will become a reference in this ever dynamic field of research.

Abstract: This paper presents a novel CNN-based architecture, referred to as Q-Net, to learn local feature descriptors that are useful for matching image patches from two different spectral bands. Given correctly matched and non-matching cross-spectral image pairs, a quadruplet network is trained to map input image patches to a common Euclidean space, regardless of the input spectral band. Our approach is inspired by the recent success of triplet networks in the visible spectrum, but adapted for cross-spectral scenarios, where, for each matching pair, there are always two possible non-matching patches: one for each spectrum. Experimental evaluations on a public cross-spectral VIS-NIR dataset shows that the proposed approach improves the state-of-the-art. Moreover, the proposed technique can also be used in mono-spectral settings, obtaining a similar performance to triplet network descriptors, but requiring less training data.

Abstract: OBJECTIVES:
The proposal and implementation of a computational framework for the quantification of structural renal damage from 99mTc-dimercaptosuccinic acid (DMSA) scans. The aim of this work is to propose, implement, and validate a computational framework for the quantification of structural renal damage from DMSA scans and in an observer-independent manner.
MATERIALS AND METHODS:
From a set of 16 pediatric DMSA-positive scans and 16 matched controls and using both expert-guided and automatic approaches, a set of image-derived quantitative indicators was computed based on the relative size, intensity and histogram distribution of the lesion. A correlation analysis was conducted in order to investigate the association of these indicators with other clinical data of interest in this scenario, including C-reactive protein (CRP), white cell count, vesicoureteral reflux, fever, relative perfusion, and the presence of renal sequelae in a 6-month follow-up DMSA scan.
RESULTS:
A fully automatic lesion detection and segmentation system was able to successfully classify DMSA-positive from negative scans (AUC=0.92, sensitivity=81% and specificity=94%). The image-computed relative size of the lesion correlated with the presence of fever and CRP levels (p<0.05), and a measurement derived from the distribution histogram of the lesion obtained significant performance results in the detection of permanent renal damage (AUC=0.86, sensitivity=100% and specificity=75%).
CONCLUSIONS:
The proposal and implementation of a computational framework for the quantification of structural renal damage from DMSA scans showed a promising potential to complement visual diagnosis and non-imaging indicators.

Abstract: The original idea of this project was to study the role of colour in the challenging task of object recognition. We started by extending previous research on colour naming showing that it is feasible to capture colour terms through parsimonious ellipsoids. Although, the results of our model exceeded state-of-the-art in two benchmark datasets, we realised that the two phenomena of metameric lights and colour constancy must be addressed prior to any further colour processing. Our investigation of metameric pairs reached the conclusion that they are infrequent in real world scenarios. Contrary to that, the illumination of a scene often changes dramatically. We addressed this issue by proposing a colour constancy model inspired by the dynamical centre-surround adaptation of neurons in the visual cortex. This was implemented through two overlapping asymmetric Gaussians whose variances and heights are adjusted according to the local contrast of pixels. We complemented this model with a generic contrast-variant pooling mechanism that inversely connect the percentage of pooled signal to the local contrast of a region. The results of our experiments on four benchmark datasets were indeed promising: the proposed model, although simple, outperformed even learning-based approaches in many cases. Encouraged by the success of our contrast-variant surround modulation, we extended this approach to detect boundaries of objects. We proposed an edge detection model based on the first derivative of the Gaussian kernel. We incorporated four types of surround: full, far, iso- and orthogonal-orientation. Furthermore, we accounted for the pooling mechanism at higher cortical areas and the shape feedback sent to lower areas. Our results in three benchmark datasets showed significant improvement over non-learning algorithms.
To summarise, we demonstrated that biologically-inspired models offer promising solutions to computer vision problems, such as, colour naming, colour constancy and edge detection. We believe that the greatest contribution of this Ph.D dissertation is modelling the concept of dynamic surround modulation that shows the significance of contrast-variant surround integration. The models proposed here are grounded on only a portion of what we know about the human visual system. Therefore, it is only natural to complement them accordingly in future works.

Abstract: Depth can complement RGB with useful cues about object volumes and scene layout. However, RGB-D image datasets are still too small for directly training deep convolutional neural networks (CNNs), in contrast to the massive monomodal RGB datasets. Previous works in RGB-D recognition typically combine two separate networks for RGB and depth data, pretrained with a large RGB dataset and then fine tuned to the respective target RGB and depth datasets. These approaches have several limitations: 1) only use low-level filters learned from RGB data, thus not being able to exploit properly depth-specific patterns, and 2) RGB and depth features are only combined at high-levels but rarely at lower-levels. In this paper, we propose a framework that leverages both knowledge acquired from large RGB datasets together with depth-specific cues learned from the limited depth data, obtaining more effective multi-source and multi-modal representations. We propose a multi-modal combination method that selects discriminative combinations of layers from the different source models and target modalities, capturing both high-level properties of the task and intrinsic low-level properties of both modalities.

Abstract: Pooling is a ubiquitous operation in image processing algorithms that allows for higher-level processes to collect relevant low-level features from a region of interest. Currently, max-pooling is one of the most commonly used operators in the computational literature. However, it can lack robustness to outliers due to the fact that it relies merely on the peak of a function. Pooling mechanisms are also present in the primate visual cortex where neurons of higher cortical areas pool signals from lower ones. The receptive fields of these neurons have been shown to vary according to the contrast by aggregating signals over a larger region in the presence of low contrast stimuli. We hypothesise that this contrast-variant-pooling mechanism can address some of the shortcomings of maxpooling. We modelled this contrast variation through a histogram clipping in which the percentage of pooled signal is inversely proportional to the local contrast of an image. We tested our hypothesis by applying it to the phenomenon of colour constancy where a number of popular algorithms utilise a max-pooling step (e.g. White-Patch, Grey-Edge and Double-Opponency). For each of these methods, we investigated the consequences of replacing their original max-pooling by the proposed contrast-variant-pooling. Our experiments on three colour constancy benchmark datasets suggest that previous results can significantly improve by adopting a contrast-variant-pooling mechanism.

Abstract: Convolutional Neural Networks (CNNs) trained for object recognition tasks present representational capabilities approaching to primate visual systems [1]. This provides a computational framework to explore how image features
are efficiently represented. Here, we dissect a trained CNN
[2] to study how color is represented. We use a classical methodology used in physiology that is measuring index of selectivity of individual neurons to specific features. We use ImageNet Dataset [20] images and synthetic versions
of them to quantify color tuning properties of artificial neurons to provide a classification of the network population.
We conclude three main levels of color representation showing some parallelisms with biological visual systems: (a) a decomposition in a circular hue space to represent single color regions with a wider hue sampling beyond the first
layer (V2), (b) the emergence of opponent low-dimensional spaces in early stages to represent color edges (V1); and (c) a strong entanglement between color and shape patterns representing object-parts (e.g. wheel of a car), objectshapes (e.g. faces) or object-surrounds configurations (e.g. blue sky surrounding an object) in deeper layers (V4 or IT).