Abstract: Recognizing object categories in real world images is a challenging problem in computer vision. The deformable part based framework is currently the most successful approach for object detection. Generally, HOG are used for image representation within the part-based framework. For action recognition, the bag-of-word framework has shown to provide promising results. Within the bag-of-words framework, local image patches are described by SIFT descriptor. Contrary to object detection and action recognition, combining color and shape has shown to provide the best performance for object and scene recognition.
In the first part of this thesis, we analyze the problem of person detection in still images. Standard person detection approaches rely on intensity based features for image representation while ignoring the color. Channel based descriptors is one of the most commonly used approaches in object recognition. This inspires us to evaluate incorporating color information using the channel based fusion approach for the task of person detection.
In the second part of the thesis, we investigate the problem of object detection in still images. Due to high dimensionality, channel based fusion increases the computational cost. Moreover, channel based fusion has been found to obtain inferior results for object category where one of the visual varies significantly. On the other hand, late fusion is known to provide improved results for a wide range of object categories. A consequence of late fusion strategy is the need of a pure color descriptor. Therefore, we propose to use Color attributes as an explicit color representation for object detection. Color attributes are compact and computationally efficient. Consequently color attributes are combined with traditional shape features providing excellent results for object detection task.
Finally, we focus on the problem of action detection and classification in still images. We investigate the potential of color for action classification and detection in still images. We also evaluate different fusion approaches for combining color and shape information for action recognition. Additionally, an analysis is performed to validate the contribution of color for action recognition. Our results clearly demonstrate that combining color and shape information significantly improve the performance of both action classification and detection in still images.

Abstract: Color constancy is usually measured by achromatic setting, asymmetric matching, or color naming paradigms, whose results are interpreted in terms of indexes and models that arguably do not capture the full complexity of the phenomenon. Here we propose a new paradigm, chromatic setting, which allows a more comprehensive characterization of color constancy through the measurement of multiple points in color space under immersive adaptation. We demonstrated its feasibility by assessing the consistency of subjects' responses over time. The paradigm was applied to two-dimensional (2-D) Mondrian stimuli under three different illuminants, and the results were used to fit a set of linear color constancy models. The use of multiple colors improved the precision of more complex linear models compared to the popular diagonal model computed from gray. Our results show that a diagonal plus translation matrix that models mechanisms other than cone gain might be best suited to explain the phenomenon. Additionally, we calculated a number of color constancy indices for several points in color space, and our results suggest that interrelations among colors are not as uniform as previously believed. To account for this variability, we developed a new structural color constancy index that takes into account the magnitude and orientation of the chromatic shift in addition to the interrelations among colors and memory effects.

Abstract: We present a novel method to auto-calibrate gaze estimators based on gaze patterns obtained from other viewers. Our method is based on the observation that the gaze patterns of humans are indicative of where a new viewer will look at [12]. When a new viewer is looking at a stimulus, we first estimate a topology of gaze points (initial gaze points). Next, these points are transformed so that they match the gaze patterns of other humans to find the correct gaze points. In a flexible uncalibrated setup with a web camera and no chin rest, the proposed method was tested on ten subjects and ten images. The method estimates the gaze points after looking at a stimulus for a few seconds with an average accuracy of 4.3 im. Although the reported performance is lower than what could be achieved with dedicated hardware or calibrated setup, the proposed method still provides a sufficient accuracy to trace the viewer attention. This is promising considering the fact that auto-calibration is done in a flexible setup , without the use of a chin rest, and based only on a few seconds of gaze initialization data. To the best of our knowledge, this is the first work to use human gaze patterns in order to auto-calibrate gaze estimators.

Abstract: The problem of training classifiers only with target data arises in many applications where non-target data are too costly, difficult to obtain, or not available at all. Several one-class classification methods have been presented to solve this problem, but most of the methods are highly sensitive to the presence of outliers in the target class. Ensemble methods have therefore been proposed as a powerful way to improve the classification performance of binary/multi-class learning algorithms by introducing diversity into classifiers.
However, their application to one-class classification has been rather limited. In
this paper, we present a new ensemble method based on a non-parametric weighted bagging strategy for one-class classification, to improve accuracy in the presence of outliers. While the standard bagging strategy assumes a uniform data distribution, the method we propose here estimates a probability density based on a forest structure of the data. This assumption allows the estimation of data distribution from the computation of simple univariate and bivariate kernel densities. Experiments using original and noisy versions of 20 different datasets show that bagging ensemble methods applied to different one-class classifiers outperform base one-class classification methods. Moreover, we show that, in noisy versions of the datasets, the non-parametric weighted bagging strategy we propose outperforms the classical bagging strategy in a statistically significant way.

Abstract: Comic books constitute an important cultural heritage asset in many countries. Digitization combined with subsequent document understanding enable direct content-based search as opposed to metadata only search (e.g. album title or author name). Few studies have been done in this direction. In this work we detail a novel approach for the automatic text localization in scanned comics book pages, an essential step towards a fully automatic comics book understanding. We focus on speech text as it is semantically important and represents the majority of the text present in comics. The approach is compared with existing methods of text localization found in the literature and results are presented.

Abstract: Detecting hands in multi-modal RGB-Depth visual data has become a challenging Computer Vision problem with several applications of interest. This task involves dealing with changes in illumination, viewpoint variations, the articulated nature of the human body, the high flexibility of the wrist articulation, and the deformability of the hand itself. In this work, we propose an accurate and efficient automatic hand detection scheme to be applied in Human-Computer Interaction (HCI) applications in which the user is seated at the desk and, thus, only the upper body is visible. Our main hypothesis is that hand landmarks remain at a nearly constant geodesic distance from an automatically located anatomical reference point.
In a given frame, the human body is segmented first in the depth image. Then, a
graph representation of the body is built in which the geodesic paths are computed from the reference point. The dense optical flow vectors on the corresponding RGB image are used to reduce ambiguities of the geodesic paths’ connectivity, allowing to eliminate false edges interconnecting different body parts. Finally, we are able to detect the position of both hands based on invariant geodesic distances and optical flow within the body region, without involving costly learning procedures.