Stefan M. Karlsson

Computing the Minimum EDF Feasible Deadline in Periodic Systems

In most real-time applications, deadlines are artifices that need to be enforced to meet different performance requirements. For example, in periodic task sets, jitter requirements can be met by assigning suitable relative deadlines and guaranteeing the feasibility of the schedule. This paper presents a method (called minD) for calculating the minimum EDF-feasible deadline of a real-time task. More precisely, given a set of periodic tasks with hard real-time requirements, which is feasible under EDF, the proposed algorithm allows computing the shortest deadline that can be assigned to an arbitrary task in the set, or to a new incoming task (periodic or aperiodic), still preserving the EDF feasibility of the new task set. The algorithm has a pseudo polynomial complexity and handles arbitrary relative deadlines, which can be less than, equal to, or greater than periods

Lip-motion events analysis and lip segmentation using optical flow

We propose an algorithm for detecting the mouth events of opening and closing. Our method is translation and rotation invariant, works at very fast speeds, and does not require segmented lips. The approach is based on a recently developed optical flow algorithm that handles the motion of linear structure in a stable and consistent way. Furthermore, we provide a semi-automatic tool for generating groundtruth segmentation of video data, also based on the optical flow algorithm used for tracking keypoints at faster than 200 frames/second. We provide groundtruth for 50 sessions of speech of the XM2VTS database [16] available for download, and the means to segment further sessions at a relatively small amount of user interaction. We use the generated groundtruth to test the proposed algorithm for detecting events, and show it to yield promising result. The semi-automatic tool will be a useful resource for researchers in need of groundtruth segmentation from video for the XM2VTS database and others.

Illuminance flow over anisotropic surfaces

A theory is presented to analyze images of anisotropic fine-scale surfaces. We investigate the estimates of illuminance flow by using structure tensors. For anisotropic surfaces, both the gradient-based and the Hessian-based tensors will yield deviations from the true illumination orientation. Our theory predicts this deviation. To show the use of this theory, an algorithm is derived that uses both tensors simultaneously to compensate for small amounts of anisotropy. Experimental results with rendered surfaces are shown to conform well to our theory.

Face Tracking Using Optical Flow

In this paper a novel face tracking approach is presented where optical flow information is incorporated into the Viola-Jones face detection algorithm. In the original algorithm from Viola and Jones face detection is static as information from previous frames is not considered. In contrast to the Viola-Jones face detector and also to other known dynamic enhancements, the proposed face tracker preserves information about near-positives. The algorithm builds a likelihood map from the intermediate results of the Viola-Jones algorithm which is extrapolated using optical flow. The objects get extracted from the likelihood map using image segmentation techniques. All steps can be computed very efficiently in real-time. The tracker is verified on the Boston Head Tracking Database showing that the proposed algorithm outperforms the standard Viola-Jones face detector.

Multiscale complex moments of the local power spectrum.

Complex moments of the local power spectrum (CMP) are investigated in a multiscale context. The multiscale CMPs are shown to approximate well the 1D angular Fourier transform of the band in question. This observation is used to derive further properties of the power spectrum in terms of texture orientations or n-folded symmetry patterns. A method is presented to approximate the power spectrum using only separable filtering in the spatial domain. Interesting implications to the Gabor decomposition are shown. The number of orientations in the filter bank is related to the order of n-folded symmetry detectable. Furthermore, the multiscale CMPs can be estimated incrementally in the spatial domain, which is both fast and reliable. Experiments on power spectrum estimation, orientation estimation, and texture segmentation are presented.

A Real-Time AdaBoost Cascade Face Tracker Based on Likelihood Map and Optical Flow

We present a novel face tracking approach where optical flow information is incorporated into a modified version of the Viola-Jones detection algorithm. In the original algorithm, detection is stat ...

Histogram of directions by the structure tensor

Many low-level features, as well as varying methods of extraction and interpretation rely on directionality analysis (for example the Hough transform, Gabor filters, SIFT descriptors and the structure tensor). The theory of the gradient based structure tensor (a.k.a. the second moment matrix) is a very well suited theoretical platform in which to analyze and explain the similarities and connections (indeed often equivalence) of supposedly different methods and features that deal with image directionality. Of special interest to this study is the SIFT descriptors (histogram of oriented gradients, HOGs). Our analysis of interrelationships of prominent directionality analysis tools offers the possibility of computation of HOGs without binning, in an algorithm of comparative time complexity.

Illuminance Flow Estimation by Regression

We investigate the estimation of illuminance flow using Histograms of Oriented Gradient features (HOGs). In a regression setting, we found for both ridge regression and support vector machines, that the optimal solution shows close resemblance to the gradient based structure tensor (also known as the second moment matrix).Theoretical results are presented showing in detail how the structure tensor and the HOGs are connected. This relation will benefit computer vision tasks such as affine invariant texture/object matching using HOGs.Several properties of HOGs are presented, among others, how many bins are required for a directionality measure, and how to estimate HOGs through spatial averaging that requires no binning.

Estimation of Illuminance Flow over Anisotropic Surfaces for Arbitrary Viewpoints

The theory of illuminance flow estimation by structure tensors is generalized for oblique viewing of anisotropic texture. Previous theory is revised using general matrix formulations and predictions are compared with results on rendered images.