Learning and Exploiting Camera Geometry for Computer Vision

This work explores the use of statistical methods in describing and estimating camera poses, as well as the information feedback loop between camera pose and object detection. Surging development in robotics and computer vision has pushed the need for algorithms that infer, understand, and utilize information about the position and orientation of the sensor platforms when observing and/or interacting with their environment.

The first contribution of this thesis is the development of a set of statistical tools for representing and estimating the uncertainty in object poses. A distribution for representing the joint uncertainty over multiple object positions and orientations is described, called the mirrored normal-Bingham distribution. This distribution generalizes both the normal distribution in Euclidean space, and the Bingham distribution on the unit hypersphere. It is shown to inherit many of the convenient properties of these special cases: it is the maximum-entropy distribution with fixed second moment, and there is a generalized Laplace approximation whose result is the mirrored normal-Bingham distribution. This distribution and approximation method are demonstrated by deriving the analytical approximation to the wrapped-normal distribution. Further, it is shown how these tools can be used to represent the uncertainty in the result of a bundle adjustment problem.

Another application of these methods is illustrated as part of a novel camera pose estimation algorithm based on object detections. The autocalibration task is formulated as a bundle adjustment problem using prior distributions over the 3D points to enforce the objects' structure and their relationship with the scene geometry. This framework is very flexible and enables the use of off-the-shelf computational tools to solve specialized autocalibration problems. Its performance is evaluated using a pedestrian detector to provide head and foot location observations, and it proves much faster and potentially more accurate than existing methods.

Finally, the information feedback loop between object detection and camera pose estimation is closed by utilizing camera pose information to improve object detection in scenarios with significant perspective warping. Methods are presented that allow the inverse perspective mapping traditionally applied to images to be applied instead to features computed from those images. For the special case of HOG-like features, which are used by many modern object detection systems, these methods are shown to provide substantial performance benefits over unadapted detectors while achieving real-time frame rates, orders of magnitude faster than comparable image warping methods.

The statistical tools and algorithms presented here are especially promising for mobile cameras, providing the ability to autocalibrate and adapt to the camera pose in real time. In addition, these methods have wide-ranging potential applications in diverse areas of computer vision, robotics, and imaging.

ImageLearn - Decoding Britain's Landscape

Abstract Ordnance Survey, our national mapping organisation, collects vast amounts of high-resolution aerial imagery covering the entirety of the country. Currently, photogrammetrists and surveyors use this to manually capture real-world objects and characteristics for a relatively small number of features. Arguably, the vast archive of imagery that we have obtained portraying the whole of Great Britain is highly underutilised and could be ‘mined’ for much more information. Over the last year the ImageLearn project has investigated the potential of "representation learning" to automatically extract relevant features from aerial imagery. Representation learning is a form of data-mining in which the feature-extractors are learned using machine-learning techniques, rather than being manually defined. At the beginning of the project we conjectured that representations learned could help with processes such as object detection and identification, change detection and social landscape regionalisation of Britain. This seminar will give an overview of the project and highlight some of our research results.

Signal subspace identification in hyperspectral imagery

Terrestrial remote sensing imagery involves the acquisition of information from the Earth's surface without physical contact with the area under study. Among the remote sensing modalities, hyperspectral imaging has recently emerged as a powerful passive technology. This technology has been widely used in the fields of urban and regional planning, water resource management, environmental monitoring, food safety, counterfeit drugs detection, oil spill and other types of chemical contamination detection, biological hazards prevention, and target detection for military and security purposes [2-9]. Hyperspectral sensors sample the reflected solar radiation from the Earth surface in the portion of the spectrum extending from the visible region through the near-infrared and mid-infrared (wavelengths between 0.3 and 2.5 µm) in hundreds of narrow (of the order of 10 nm) contiguous bands [10]. This high spectral resolution can be used for object detection and for discriminating between different objects based on their spectral xharacteristics [6]. However, this huge spectral resolution yields large amounts of data to be processed. For example, the Airbone Visible/Infrared Imaging Spectrometer (AVIRIS) [11] collects a 512 (along track) X 614 (across track) X 224 (bands) X 12 (bits) data cube in 5 s, corresponding to about 140 MBs. Similar data collection ratios are achieved by other spectrometers [12]. Such huge data volumes put stringent requirements on communications, storage, and processing. The problem of signal sbspace identification of hyperspectral data represents a crucial first step in many hypersctral processing algorithms such as target detection, change detection, classification, and unmixing. The identification of this subspace enables a correct dimensionality reduction (DR) yelding gains in data storage and retrieval and in computational time and complexity. Additionally, DR may also improve algorithms performance since it reduce data dimensionality without losses in the useful signal components. The computation of statistical estimates is a relevant example of the advantages of DR, since the number of samples required to obtain accurate estimates increases drastically with the dimmensionality of the data (Hughes phnomenon) [13].

Hierarchically grouped 2D local features applied to edge contour localisation

One of the most significant research topics in computer vision is object detection. Most of the reported object detection results localise the detected object within a bounding box, but do not explicitly label the edge contours of the object. Since object contours provide a fundamental diagnostic of object shape, some researchers have initiated work on linear contour feature representations for object detection and localisation. However, linear contour feature-based localisation is highly dependent on the performance of linear contour detection within natural images, and this can be perturbed significantly by a cluttered background. In addition, the conventional approach to achieving rotation-invariant features is to rotate the feature receptive field to align with the local dominant orientation before computing the feature representation. Grid resampling after rotation adds extra computational cost and increases the total time consumption for computing the feature descriptor. Though it is not an expensive process if using current computers, it is appreciated that if each step of the implementation is faster to compute especially when the number of local features is increasing and the application is implemented on resource limited ”smart devices”, such as mobile phones, in real-time. Motivated by the above issues, a 2D object localisation system is proposed in this thesis that matches features of edge contour points, which is an alternative method that takes advantage of the shape information for object localisation. This is inspired by edge contour points comprising the basic components of shape contours. In addition, edge point detection is usually simpler to achieve than linear edge contour detection. Therefore, the proposed localization system could avoid the need for linear contour detection and reduce the pathological disruption from the image background. Moreover, since natural images usually comprise many more edge contour points than interest points (i.e. corner points), we also propose new methods to generate rotation-invariant local feature descriptors without pre-rotating the feature receptive field to improve the computational efficiency of the whole system. In detail, the 2D object localisation system is achieved by matching edge contour points features in a constrained search area based on the initial pose-estimate produced by a prior object detection process. The local feature descriptor obtains rotation invariance by making use of rotational symmetry of the hexagonal structure. Therefore, a set of local feature descriptors is proposed based on the hierarchically hexagonal grouping structure. Ultimately, the 2D object localisation system achieves a very promising performance based on matching the proposed features of edge contour points with the mean correct labelling rate of the edge contour points 0.8654 and the mean false labelling rate 0.0314 applied on the data from Amsterdam Library of Object Images (ALOI). Furthermore, the proposed descriptors are evaluated by comparing to the state-of-the-art descriptors and achieve competitive performances in terms of pose estimate with around half-pixel pose error.

Active Foreground Region Extraction and Tracking for Sports Video Annotation

Automatic video segmentation plays a vital role in sports videos annotation. This paper presents a fully automatic and computationally efficient algorithm for analysis of sports videos. Various methods of automatic shot boundary detection have been proposed to perform automatic video segmentation. These investigations mainly concentrate on detecting fades and dissolves for fast processing of the entire video scene without providing any additional feedback on object relativity within the shots. The goal of the proposed method is to identify regions that perform certain activities in a scene. The model uses some low-level feature video processing algorithms to extract the shot boundaries from a video scene and to identify dominant colours within these boundaries. An object classification method is used for clustering the seed distributions of the dominant colours to homogeneous regions. Using a simple tracking method a classification of these regions to active or static is performed. The efficiency of the proposed framework is demonstrated over a standard video benchmark with numerous types of sport events and the experimental results show that our algorithm can be used with high accuracy for automatic annotation of active regions for sport videos.

Context Models for Understanding Images and Videos

A computer vision system that has to interact in natural language needs to understand the visual appearance of interactions between objects along with the appearance of objects themselves. Relationships between objects are frequently mentioned in queries of tasks like semantic image retrieval, image captioning, visual question answering and natural language object detection. Hence, it is essential to model context between objects for solving these tasks. In the first part of this thesis, we present a technique for detecting an object mentioned in a natural language query. Specifically, we work with referring expressions which are sentences that identify a particular object instance in an image. In many referring expressions, an object is described in relation to another object using prepositions, comparative adjectives, action verbs etc. Our proposed technique can identify both the referred object and the context object mentioned in such expressions. Context is also useful for incrementally understanding scenes and videos. In the second part of this thesis, we propose techniques for searching for objects in an image and events in a video. Our proposed incremental algorithms use the context from previously explored regions to prioritize the regions to explore next. The advantage of incremental understanding is restricting the amount of computation time and/or resources spent for various detection tasks. Our first proposed technique shows how to learn context in indoor scenes in an implicit manner and use it for searching for objects. The second technique shows how explicitly written context rules of one-on-one basketball can be used to sequentially detect events in a game.

Comparison of Computer-Based and Optical Face Recognition Paradigms

The main objectives of this thesis are to validate an improved principal components analysis (IPCA) algorithm on images; designing and simulating a digital model for image compression, face recognition and image detection by using a principal components analysis (PCA) algorithm and the IPCA algorithm; designing and simulating an optical model for face recognition and object detection by using the joint transform correlator (JTC); establishing detection and recognition thresholds for each model; comparing between the performance of the PCA algorithm and the performance of the IPCA algorithm in compression, recognition and, detection; and comparing between the performance of the digital model and the performance of the optical model in recognition and detection. The MATLAB © software was used for simulating the models. PCA is a technique used for identifying patterns in data and representing the data in order to highlight any similarities or differences. The identification of patterns in data of high dimensions (more than three dimensions) is too difficult because the graphical representation of data is impossible. Therefore, PCA is a powerful method for analyzing data. IPCA is another statistical tool for identifying patterns in data. It uses information theory for improving PCA. The joint transform correlator (JTC) is an optical correlator used for synthesizing a frequency plane filter for coherent optical systems. The IPCA algorithm, in general, behaves better than the PCA algorithm in the most of the applications. It is better than the PCA algorithm in image compression because it obtains higher compression, more accurate reconstruction, and faster processing speed with acceptable errors; in addition, it is better than the PCA algorithm in real-time image detection due to the fact that it achieves the smallest error rate as well as remarkable speed. On the other hand, the PCA algorithm performs better than the IPCA algorithm in face recognition because it offers an acceptable error rate, easy calculation, and a reasonable speed. Finally, in detection and recognition, the performance of the digital model is better than the performance of the optical model.

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.

Robotic Sensing and Manipulation of Deformable Linear Objects with Learning-based methods

Nowadays robotic applications are widespread and most of the manipulation tasks are efficiently solved. However, Deformable-Objects (DOs) still represent a huge limitation for robots. The main difficulty in DOs manipulation is dealing with the shape and dynamics uncertainties, which prevents the use of model-based approaches (since they are excessively computationally complex) and makes sensory data difficult to interpret. This thesis reports the research activities aimed to address some applications in robotic manipulation and sensing of Deformable-Linear-Objects (DLOs), with particular focus to electric wires. In all the works, a significant effort was made in the study of an effective strategy for analyzing sensory signals with various machine learning algorithms. In the former part of the document, the main focus concerns the wire terminals, i.e. detection, grasping, and insertion. First, a pipeline that integrates vision and tactile sensing is developed, then further improvements are proposed for each module. A novel procedure is proposed to gather and label massive amounts of training images for object detection with minimal human intervention. Together with this strategy, we extend a generic object detector based on Convolutional-Neural-Networks for orientation prediction. The insertion task is also extended by developing a closed-loop control capable to guide the insertion of a longer and curved segment of wire through a hole, where the contact forces are estimated by means of a Recurrent-Neural-Network. In the latter part of the thesis, the interest shifts to the DLO shape. Robotic reshaping of a DLO is addressed by means of a sequence of pick-and-place primitives, while a decision making process driven by visual data learns the optimal grasping locations exploiting Deep Q-learning and finds the best releasing point. The success of the solution leverages on a reliable interpretation of the DLO shape. For this reason, further developments are made on the visual segmentation.

Deep learning models for the automatic analysis of volleyball plays

Collecting and analysing data is an important element in any field of human activity and research. Even in sports, collecting and analyzing statistical data is attracting a growing interest. Some exemplar use cases are: improvement of technical/tactical aspects for team coaches, definition of game strategies based on the opposite team play or evaluation of the performance of players. Other advantages are related to taking more precise and impartial judgment in referee decisions: a wrong decision can change the outcomes of important matches. Finally, it can be useful to provide better representations and graphic effects that make the game more engaging for the audience during the match. Nowadays it is possible to delegate this type of task to automatic software systems that can use cameras or even hardware sensors to collect images or data and process them. One of the most efficient methods to collect data is to process the video images of the sporting event through mixed techniques concerning machine learning applied to computer vision. As in other domains in which computer vision can be applied, the main tasks in sports are related to object detection, player tracking, and to the pose estimation of athletes. The goal of the present thesis is to apply different models of CNNs to analyze volleyball matches. Starting from video frames of a volleyball match, we reproduce a bird's eye view of the playing court where all the players are projected, reporting also for each player the type of action she/he is performing.

Tecniche di deep learning per il monitoraggio del processo nell’industria dello pneumatico

Il seguente elaborato affronta l'implementazione di un algoritmo che affronta un problema di controllo di processo in ambito industriale utilizzando algoritmi di object detection. Infatti, il progetto concordato con il professore Di Stefano si è svolto in collaborazione con l’azienda Pirelli, nell’ambito della produzione di pneumatici. Lo scopo dell'algoritmo implementato è di verificare il preciso orientamento di elementi grafici della copertura, utilizzati dalle case automobilistiche per equipaggiare correttamente le vetture. In particolare, si devono individuare delle scritte sul battistrada della copertura e identificarne la posizione rispetto ad altri elementi fissati su di essa. La tesi affronta questo task in due parti distinte: la prima consiste nel training di algoritmi di deep learning per il riconoscimento degli elementi grafici e del battistrada, la seconda è un decisore che opera a valle del primo sistema utilizzando gli output delle reti allenate.

Collection and real-time analysis of images from aircraft to support civil protection operational activities

This report describes the realization of a system, in which an object detection model will be implemented, whose aim is to detect the presence of people in images. This system could be used for several applications: for example, it could be carried on board an aircraft or a drone. In this case, the system is designed in such a way that it can be mounted on light/medium weight helicopters, helping the operator to find people in emergency situations. In the first chapter the use of helicopters for civil protection is analysed and applications similar to this case study are listed. The second chapter describes the choice of the hardware devices that have been used to implement a prototype of a system to collect, analyse and display images. At first, the PC necessary to process the images was chosen, based on the characteristics of the algorithms that are necessary to run the analysis. In the further, a camera that could be compatible with the PC was selected. Finally, the battery pack was chosen taking into account the electrical consumption of the devices. The third chapter illustrates the algorithms used for image analysis. In the fourth, some of the requirements listed in the regulations that must be taken into account for carrying on board all the devices have been briefly analysed. In the fifth chapter the activity of design and modelling, with the CAD Solidworks, the devices and a prototype of a case that will house them is described. The sixth chapter discusses the additive manufacturing, since the case was printed exploiting this technology. In the seventh chapter, part of the tests that must be carried out on the equipment to certificate it have been analysed, and some simulations have been carried out. In the eighth chapter the results obtained once loaded the object detection model on a hardware for image analyses were showed. In the ninth chapter, conclusions and future applications were discussed.

Analisi di video con visione binoculare

Nel TCR - Termina container Ravenna, è importante che nel momento di scarico del container sul camion non siano presenti persone nell’area. In questo elaborato si descrive la realizzazione e il funzionamento di un sistema di allarme automatico, in grado di rilevare persone ed eventualmente interrompere la procedura di scarico del container. Tale sistema si basa sulla tecnica della object segmentation tramite rimozione dello sfondo, a cui viene affiancata una classificazione e rimozione delle eventuali ombre con un metodo cromatico. Inoltre viene identificata la possibile testa di una persona e avendo a disposizione due telecamere, si mette in atto una visione binoculare per calcolarne l’altezza. Infine, viene presa in considerazione anche la dinamica del sistema, per cui la classificazione di una persona si può basare sulla grandezza, altezza e velocità dell’oggetto individuato.

Robotic grasping: State of the art and modular real-world implementation

Robotic Grasping is an important research topic in robotics since for robots to attain more general-purpose utility, grasping is a necessary skill, but very challenging to master. In general the robots may use their perception abilities like an image from a camera to identify grasps for a given object usually unknown. A grasp describes how a robotic end-effector need to be positioned to securely grab an object and successfully lift it without lost it, at the moment state of the arts solutions are still far behind humans. In the last 5–10 years, deep learning methods take the scene to overcome classical problem like the arduous and time-consuming approach to form a task-specific algorithm analytically. In this thesis are present the progress and the approaches in the robotic grasping field and the potential of the deep learning methods in robotic grasping. Based on that, an implementation of a Convolutional Neural Network (CNN) as a starting point for generation of a grasp pose from camera view has been implemented inside a ROS environment. The developed technologies have been integrated into a pick-and-place application for a Panda robot from Franka Emika. The application includes various features related to object detection and selection. Additionally, the features have been kept as generic as possible to allow for easy replacement or removal if needed, without losing time for improvement or new testing.

Land use / land cover change detection: an object oriented approach, Münster, Germany

Dissertation submitted in partial fulfilment of the requirements for the Degree of Master of Science in Geospatial Technologies