Hand detection on images based on deformable part models and additional features

Hand detection on images has important applications on person activities recognition. This thesis focuses on PASCAL Visual Object Classes (VOC) system for hand detection. VOC has become a popular system for object detection, based on twenty common objects, and has been released with a successful deformable parts model in VOC2007. A hand detection on an image is made when the system gets a bounding box which overlaps with at least 50% of any ground truth bounding box for a hand on the image. The initial average precision of this detector is around 0.215 compared with a state-of-art of 0.104; however, color and frequency features for detected bounding boxes contain important information for re-scoring, and the average precision can be improved to 0.218 with these features. Results show that these features help on getting higher precision for low recall, even though the average precision is similar.

Detecção de região saliente em imagens usando dissimilaridade de cor e amostragem por pixeis aleatórios

The purpose of this work is to demonstrate and to assess a simple algorithm for automatic estimation of the most salient region in an image, that have possible application in computer vision. The algorithm uses the connection between color dissimilarities in the image and the image’s most salient region. The algorithm also avoids using image priors. Pixel dissimilarity is an informal function of the distance of a specific pixel’s color to other pixels’ colors in an image. We examine the relation between pixel color dissimilarity and salient region detection on the MSRA1K image dataset. We propose a simple algorithm for salient region detection through random pixel color dissimilarity. We define dissimilarity by accumulating the distance between each pixel and a sample of n other random pixels, in the CIELAB color space. An important result is that random dissimilarity between each pixel and just another pixel (n = 1) is enough to create adequate saliency maps when combined with median filter, with competitive average performance if compared with other related methods in the saliency detection research field. The assessment was performed by means of precision-recall curves. This idea is inspired on the human attention mechanism that is able to choose few specific regions to focus on, a biological system that the computer vision community aims to emulate. We also review some of the history on this topic of selective attention.

Integrated visual perception architecture for robotic clothes perception and manipulation

This thesis proposes a generic visual perception architecture for robotic clothes perception and manipulation. This proposed architecture is fully integrated with a stereo vision system and a dual-arm robot and is able to perform a number of autonomous laundering tasks. Clothes perception and manipulation is a novel research topic in robotics and has experienced rapid development in recent years. Compared to the task of perceiving and manipulating rigid objects, clothes perception and manipulation poses a greater challenge. This can be attributed to two reasons: firstly, deformable clothing requires precise (high-acuity) visual perception and dexterous manipulation; secondly, as clothing approximates a non-rigid 2-manifold in 3-space, that can adopt a quasi-infinite configuration space, the potential variability in the appearance of clothing items makes them difficult to understand, identify uniquely, and interact with by machine. From an applications perspective, and as part of EU CloPeMa project, the integrated visual perception architecture refines a pre-existing clothing manipulation pipeline by completing pre-wash clothes (category) sorting (using single-shot or interactive perception for garment categorisation and manipulation) and post-wash dual-arm flattening. To the best of the author’s knowledge, as investigated in this thesis, the autonomous clothing perception and manipulation solutions presented here were first proposed and reported by the author. All of the reported robot demonstrations in this work follow a perception-manipulation method- ology where visual and tactile feedback (in the form of surface wrinkledness captured by the high accuracy depth sensor i.e. CloPeMa stereo head or the predictive confidence modelled by Gaussian Processing) serve as the halting criteria in the flattening and sorting tasks, respectively. From scientific perspective, the proposed visual perception architecture addresses the above challenges by parsing and grouping 3D clothing configurations hierarchically from low-level curvatures, through mid-level surface shape representations (providing topological descriptions and 3D texture representations), to high-level semantic structures and statistical descriptions. A range of visual features such as Shape Index, Surface Topologies Analysis and Local Binary Patterns have been adapted within this work to parse clothing surfaces and textures and several novel features have been devised, including B-Spline Patches with Locality-Constrained Linear coding, and Topology Spatial Distance to describe and quantify generic landmarks (wrinkles and folds). The essence of this proposed architecture comprises 3D generic surface parsing and interpretation, which is critical to underpinning a number of laundering tasks and has the potential to be extended to other rigid and non-rigid object perception and manipulation tasks. The experimental results presented in this thesis demonstrate that: firstly, the proposed grasp- ing approach achieves on-average 84.7% accuracy; secondly, the proposed flattening approach is able to flatten towels, t-shirts and pants (shorts) within 9 iterations on-average; thirdly, the proposed clothes recognition pipeline can recognise clothes categories from highly wrinkled configurations and advances the state-of-the-art by 36% in terms of classification accuracy, achieving an 83.2% true-positive classification rate when discriminating between five categories of clothes; finally the Gaussian Process based interactive perception approach exhibits a substantial improvement over single-shot perception. Accordingly, this thesis has advanced the state-of-the-art of robot clothes perception and manipulation.

EACS: Effective Avoidance Combination Strategy

International audience

Contributions to 3D Data Registration and Representation

Nowadays, new computers generation provides a high performance that enables to build computationally expensive computer vision applications applied to mobile robotics. Building a map of the environment is a common task of a robot and is an essential part to allow the robots to move through these environments. Traditionally, mobile robots used a combination of several sensors from different technologies. Lasers, sonars and contact sensors have been typically used in any mobile robotic architecture, however color cameras are an important sensor due to we want the robots to use the same information that humans to sense and move through the different environments. Color cameras are cheap and flexible but a lot of work need to be done to give robots enough visual understanding of the scenes. Computer vision algorithms are computational complex problems but nowadays robots have access to different and powerful architectures that can be used for mobile robotics purposes. The advent of low-cost RGB-D sensors like Microsoft Kinect which provide 3D colored point clouds at high frame rates made the computer vision even more relevant in the mobile robotics field. The combination of visual and 3D data allows the systems to use both computer vision and 3D processing and therefore to be aware of more details of the surrounding environment. The research described in this thesis was motivated by the need of scene mapping. Being aware of the surrounding environment is a key feature in many mobile robotics applications from simple robotic navigation to complex surveillance applications. In addition, the acquisition of a 3D model of the scenes is useful in many areas as video games scene modeling where well-known places are reconstructed and added to game systems or advertising where once you get the 3D model of one room the system can add furniture pieces using augmented reality techniques. In this thesis we perform an experimental study of the state-of-the-art registration methods to find which one fits better to our scene mapping purposes. Different methods are tested and analyzed on different scene distributions of visual and geometry appearance. In addition, this thesis proposes two methods for 3d data compression and representation of 3D maps. Our 3D representation proposal is based on the use of Growing Neural Gas (GNG) method. This Self-Organizing Maps (SOMs) has been successfully used for clustering, pattern recognition and topology representation of various kind of data. Until now, Self-Organizing Maps have been primarily computed offline and their application in 3D data has mainly focused on free noise models without considering time constraints. Self-organising neural models have the ability to provide a good representation of the input space. In particular, the Growing Neural Gas (GNG) is a suitable model because of its flexibility, rapid adaptation and excellent quality of representation. However, this type of learning is time consuming, specially for high-dimensional input data. Since real applications often work under time constraints, it is necessary to adapt the learning process in order to complete it in a predefined time. This thesis proposes a hardware implementation leveraging the computing power of modern GPUs which takes advantage of a new paradigm coined as General-Purpose Computing on Graphics Processing Units (GPGPU). Our proposed geometrical 3D compression method seeks to reduce the 3D information using plane detection as basic structure to compress the data. This is due to our target environments are man-made and therefore there are a lot of points that belong to a plane surface. Our proposed method is able to get good compression results in those man-made scenarios. The detected and compressed planes can be also used in other applications as surface reconstruction or plane-based registration algorithms. Finally, we have also demonstrated the goodness of the GPU technologies getting a high performance implementation of a CAD/CAM common technique called Virtual Digitizing.

Generating Discriminative Objective Proposals via Submodular Ranking

Object recognition has long been a core problem in computer vision. To improve object spatial support and speed up object localization for object recognition, generating high-quality category-independent object proposals as the input for object recognition system has drawn attention recently. Given an image, we generate a limited number of high-quality and category-independent object proposals in advance and used as inputs for many computer vision tasks. We present an efficient dictionary-based model for image classification task. We further extend the work to a discriminative dictionary learning method for tensor sparse coding. In the first part, a multi-scale greedy-based object proposal generation approach is presented. Based on the multi-scale nature of objects in images, our approach is built on top of a hierarchical segmentation. We first identify the representative and diverse exemplar clusters within each scale. Object proposals are obtained by selecting a subset from the multi-scale segment pool via maximizing a submodular objective function, which consists of a weighted coverage term, a single-scale diversity term and a multi-scale reward term. The weighted coverage term forces the selected set of object proposals to be representative and compact; the single-scale diversity term encourages choosing segments from different exemplar clusters so that they will cover as many object patterns as possible; the multi-scale reward term encourages the selected proposals to be discriminative and selected from multiple layers generated by the hierarchical image segmentation. The experimental results on the Berkeley Segmentation Dataset and PASCAL VOC2012 segmentation dataset demonstrate the accuracy and efficiency of our object proposal model. Additionally, we validate our object proposals in simultaneous segmentation and detection and outperform the state-of-art performance. To classify the object in the image, we design a discriminative, structural low-rank framework for image classification. We use a supervised learning method to construct a discriminative and reconstructive dictionary. By introducing an ideal regularization term, we perform low-rank matrix recovery for contaminated training data from all categories simultaneously without losing structural information. A discriminative low-rank representation for images with respect to the constructed dictionary is obtained. With semantic structure information and strong identification capability, this representation is good for classification tasks even using a simple linear multi-classifier.

Object detection and recognition in complex scenes

Dissertação de Mestrado, Engenharia Informática, Faculdade de Ciências e Tecnologia, Universidade do Algarve, 2014

Quantum of Vision

Visual inputs to artificial and biological visual systems are often quantized: cameras accumulate photons from the visual world, and the brain receives action potentials from visual sensory neurons. Collecting more information quanta leads to a longer acquisition time and better performance. In many visual tasks, collecting a small number of quanta is sufficient to solve the task well. The ability to determine the right number of quanta is pivotal in situations where visual information is costly to obtain, such as photon-starved or time-critical environments. In these situations, conventional vision systems that always collect a fixed and large amount of information are infeasible. I develop a framework that judiciously determines the number of information quanta to observe based on the cost of observation and the requirement for accuracy. The framework implements the optimal speed versus accuracy tradeoff when two assumptions are met, namely that the task is fully specified probabilistically and constant over time. I also extend the framework to address scenarios that violate the assumptions. I deploy the framework to three recognition tasks: visual search (where both assumptions are satisfied), scotopic visual recognition (where the model is not specified), and visual discrimination with unknown stimulus onset (where the model is dynamic over time). Scotopic classification experiments suggest that the framework leads to dramatic improvement in photon-efficiency compared to conventional computer vision algorithms. Human psychophysics experiments confirmed that the framework provides a parsimonious and versatile explanation for human behavior under time pressure in both static and dynamic environments.

Guided Bus Rapid Transit

Electrical Bus Rapid Transit (eBRT) is a charging electrical public transport which brings a clean, high performance, and affordable cost alternative from the conventional traffic vehicles which work with combustion and hybrid technology. These buses charge the battery in every bus stop to arrive at the next station. But, this charging system needs an appropriate infrastructure called pantograph, and it requires a high precision bus location to maintain battery lifetime, energy saving and charging time. To overcome this issue Vicomtech and Datik has planned a project based on computer vision to help to the driver to locate the vehicle in the correct place. In this document, we present a mono camera bus driver guided fast algorithm because these vehicles embedded computers do not support high computation and precision operations. In addition to the frequent lane sign, there are more accurate geometric beacons painted on the road to bring metric information to the vision system. This method uses segmentation to binarize the image discriminating the background space. Besides it detects, tracks and counts different lane mark contours in addition to classify each special painted mark. Besides it does not need any calibration task to calculate longitudinal and cross distances because we know the lane mark sizes.

Improving surgery operations by means of cloud systems and distributed user interfaces

Surgical interventions are usually performed in an operation room; however, access to the information by the medical team members during the intervention is limited. While in conversations with the medical staff, we observed that they attach significant importance to the improvement of the information and communication direct access by queries during the process in real time. It is due to the fact that the procedure is rather slow and there is lack of interaction with the systems in the operation room. These systems can be integrated on the Cloud adding new functionalities to the existing systems the medical expedients are processed. Therefore, such a communication system needs to be built upon the information and interaction access specifically designed and developed to aid the medical specialists. Copyright 2014 ACM.

Improving user-insurance communication on accident reports

This paper presents an easy to use methodology and system for insurance companies targeting at managing traffic accidents reports process. The main objective is to facilitate and accelerate the process of creating and finalizing the necessary accident reports in cases without mortal victims involved. The diverse entities participating in the process from the moment an accident occurs until the related final actions needed are included. Nowadays, this market is limited to the consulting platforms offered by the insurance companies. Copyright 2014 ACM.

Scene Mapping and Understanding by Robotic Vision

The first mechanical Automaton concept was found in a Chinese text written in the 3rd century BC, while Computer Vision was born in the late 1960s. Therefore, visual perception applied to machines (i.e. the Machine Vision) is a young and exciting alliance. When robots came in, the new field of Robotic Vision was born, and these terms began to be erroneously interchanged. In short, we can say that Machine Vision is an engineering domain, which concern the industrial use of Vision. The Robotic Vision, instead, is a research field that tries to incorporate robotics aspects in computer vision algorithms. Visual Servoing, for example, is one of the problems that cannot be solved by computer vision only. Accordingly, a large part of this work deals with boosting popular Computer Vision techniques by exploiting robotics: e.g. the use of kinematics to localize a vision sensor, mounted as the robot end-effector. The remainder of this work is dedicated to the counterparty, i.e. the use of computer vision to solve real robotic problems like grasping objects or navigate avoiding obstacles. Will be presented a brief survey about mapping data structures most widely used in robotics along with SkiMap, a novel sparse data structure created both for robotic mapping and as a general purpose 3D spatial index. Thus, several approaches to implement Object Detection and Manipulation, by exploiting the aforementioned mapping strategies, will be proposed, along with a completely new Machine Teaching facility in order to simply the training procedure of modern Deep Learning networks.

Concepts and methods for efficient decentralized learning in federated settings

Deep Neural Networks (DNNs) have revolutionized a wide range of applications beyond traditional machine learning and artificial intelligence fields, e.g., computer vision, healthcare, natural language processing and others. At the same time, edge devices have become central in our society, generating an unprecedented amount of data which could be used to train data-hungry models such as DNNs. However, the potentially sensitive or confidential nature of gathered data poses privacy concerns when storing and processing them in centralized locations. To this purpose, decentralized learning decouples model training from the need of directly accessing raw data, by alternating on-device training and periodic communications. The ability of distilling knowledge from decentralized data, however, comes at the cost of facing more challenging learning settings, such as coping with heterogeneous hardware and network connectivity, statistical diversity of data, and ensuring verifiable privacy guarantees. This Thesis proposes an extensive overview of decentralized learning literature, including a novel taxonomy and a detailed description of the most relevant system-level contributions in the related literature for privacy, communication efficiency, data and system heterogeneity, and poisoning defense. Next, this Thesis presents the design of an original solution to tackle communication efficiency and system heterogeneity, and empirically evaluates it on federated settings. For communication efficiency, an original method, specifically designed for Convolutional Neural Networks, is also described and evaluated against the state-of-the-art. Furthermore, this Thesis provides an in-depth review of recently proposed methods to tackle the performance degradation introduced by data heterogeneity, followed by empirical evaluations on challenging data distributions, highlighting strengths and possible weaknesses of the considered solutions. Finally, this Thesis presents a novel perspective on the usage of Knowledge Distillation as a mean for optimizing decentralized learning systems in settings characterized by data heterogeneity or system heterogeneity. Our vision on relevant future research directions close the manuscript.

Computational methods applied to undeciphered scripts from the Aegean and Cyprus: Linear A and Cypro-Minoan

The study of ancient, undeciphered scripts presents unique challenges, that depend both on the nature of the problem and on the peculiarities of each writing system. In this thesis, I present two computational approaches that are tailored to two different tasks and writing systems. The first of these methods is aimed at the decipherment of the Linear A afraction signs, in order to discover their numerical values. This is achieved with a combination of constraint programming, ad-hoc metrics and paleographic considerations. The second main contribution of this thesis regards the creation of an unsupervised deep learning model which uses drawings of signs from ancient writing system to learn to distinguish different graphemes in the vector space. This system, which is based on techniques used in the field of computer vision, is adapted to the study of ancient writing systems by incorporating information about sequences in the model, mirroring what is often done in natural language processing. In order to develop this model, the Cypriot Greek Syllabary is used as a target, since this is a deciphered writing system. Finally, this unsupervised model is adapted to the undeciphered Cypro-Minoan and it is used to answer open questions about this script. In particular, by reconstructing multiple allographs that are not agreed upon by paleographers, it supports the idea that Cypro-Minoan is a single script and not a collection of three script like it was proposed in the literature. These results on two different tasks shows that computational methods can be applied to undeciphered scripts, despite the relatively low amount of available data, paving the way for further advancement in paleography using these methods.

Continual learning on nano drones

Miniaturized flying robotic platforms, called nano-drones, have the potential to revolutionize the autonomous robots industry sector thanks to their very small form factor. The nano-drones’ limited payload only allows for a sub-100mW microcontroller unit for the on-board computations. Therefore, traditional computer vision and control algorithms are too computationally expensive to be executed on board these palm-sized robots, and we are forced to rely on artificial intelligence to trade off accuracy in favor of lightweight pipelines for autonomous tasks. However, relying on deep learning exposes us to the problem of generalization since the deployment scenario of a convolutional neural network (CNN) is often composed by different visual cues and different features from those learned during training, leading to poor inference performances. Our objective is to develop and deploy and adaptation algorithm, based on the concept of latent replays, that would allow us to fine-tune a CNN to work in new and diverse deployment scenarios. To do so we start from an existing model for visual human pose estimation, called PULPFrontnet, which is used to identify the pose of a human subject in space through its 4 output variables, and we present the design of our novel adaptation algorithm, which features automatic data gathering and labeling and on-device deployment. We therefore showcase the ability of our algorithm to adapt PULP-Frontnet to new deployment scenarios, improving the R2 scores of the four network outputs, with respect to an unknown environment, from approximately [−0.2, 0.4, 0.0,−0.7] to [0.25, 0.45, 0.2, 0.1]. Finally we demonstrate how it is possible to fine-tune our neural network in real time (i.e., under 76 seconds), using the target parallel ultra-low power GAP 8 System-on-Chip on board the nano-drone, and we show how all adaptation operations can take place using less than 2mWh of energy, a small fraction of the available battery power.