In this thesis, we study two closely related directions: robustness and generalization in modern deep learning. Deep learning models based on empirical risk minimization are known to be often non-robust to small, worst-case perturbations known as adversari ...
The Universal Serial Bus (USB) connects external devices to a host. This interface exposes the OS kernels and device drivers to attacks by malicious devices. Unfortunately, kernels and drivers were developed under a security model that implicitly trusts co ...
We consider the robust linear regression problem in the online setting where we have access to the data in a streaming manner, one data point after the other. More specifically, for a true parameter , we consider the corrupted Gaussian linear model $ y=\la ...
Multiple object tracking is a crucial Computer Vision Task. It aims at locating objects of interest in the image sequences, maintaining their identities, and identifying their trajectories over time. A large portion of current research focuses on tracking ...
We propose a non-parametric regression method that does not rely on the structure of the ground-truth, but only on its regularity properties. The methodology can be readily used for learning surrogate models of nonlinear dynamical systems from data, while ...
In our everyday life we interact with the surrounding environment using our hands. A main focus of recent research has been to bring such interaction to virtual objects, such as the ones projected in virtual reality devices, or super-imposed as holograms i ...
Accurate and robust 3D head pose estimation is important for face related analysis. Though high accuracy has been achieved by previous works based on 3D morphable model (3DMM), their performance drops with extreme head poses because such models usually onl ...
This thesis addresses the problem of recovering the 3-D shape of a deformable object in single images, or image sequences acquired by a monocular video camera, given that a 3-D template shape and a template image of the object are available. While being a ...
Recent advances in Computer Vision are changing our way of living and enabling new applications for both leisure and professional use. Regrettably, in many industrial domains the spread of state-of-the-art technologies is made challenging by the abundance ...
Interest in deep probabilistic graphical models has increased in recent years, due to their state-of-the-art perfor- mance on many machine learning applications. Such models are typically trained with the stochastic gradient method, which can take a signif ...
Alternating direction method of multipliers (ADMM) is a form of augmented Lagrangian optimisation algorithm that found its place in many new applications in recent years. This paper explores a possibility for an upgrade of the ADMM by extrapolation-based a ...
The contribution of this paper is two-fold. First, we show how to extend the ESM algorithm to handle motion blur in 3D object tracking. ESM is a powerful algorithm for template matching-based tracking, but it can fail under motion blur. We introduce an ima ...
In the context of retinal microsurgery, visual tracking of instruments is a key component of robotics assistance. The difficulty of the task and major reason why most existing strategies fail on {\it in-vivo} image sequences lies in the fact that complex a ...
This paper examines the computational complexity certification of the fast gradient method for the solution of the dual of a parametric con- vex program. To this end, a lower iteration bound is derived such that for all parameters from a compact set a solu ...
We propose a real-time solution for modeling and tracking multiple 3D objects in unknown environments for Augmented Reality. The proposed solution consists of both scalable tracking and interactive modeling. Our contribution is twofold: First, we show how ...
Spectral reflection prediction models, although effective, are impractical for certain industrial applications such as self-calibrating devices and online monitoring because of the requirements imposed by their calibration. The idea emerged to make the cal ...
Structural health monitoring (SHM) has the potential to provide quantitative and reliable data on the real condition of structures, observe the evolution of their behaviour and detect degradation This paper presents two methodologies for model-free data in ...
The aim of this thesis is to build a system able to automatically and robustly track human motion in 3–D starting from monocular input. To this end two approaches are introduced, which tackle two different types of motion: The first is useful to analyze ac ...
We combine detection and tracking techniques to achieve robust 3D motion recovery of people seen from arbitrary viewpoints by a single and potentially moving camera. We rely on detecting key postures, which can be done reliably, using a motion model to inf ...
Institute of Electrical and Electronics Engineers2010
The contribution of this paper is two-fold. First, we show how to extend the ESM algorithm to handle motion blur in 3D object track- ing. ESM is a powerful algorithm for template matching-based tracking, but it can fail under motion blur. We introduce an i ...
