Abstract: In the context of human action recognition using skeleton data, the 3D trajectories of joint points may be considered as multi-dimensional time series. The traditional recognition technique in the literature is based on time series dis(similarity) measures (such as Dynamic Time Warping). For these general dis(similarity) measures, k-nearest neighbor algorithms are a natural choice. However, k-NN classifiers are known to be sensitive to noise and outliers. In this paper, a new class of Support Vector Machine that is applicable to trajectory classification, such as action recognition, is developed by incorporating an efficient time-series distances measure into the kernel function. More specifically, the derivative of Dynamic Time Warping (DTW) distance measure is employed as the SVM kernel. In addition, the pairwise proximity learning strategy is utilized in order to make use of non-positive semi-definite (PSD) kernels in the SVM formulation. The recognition results of the proposed technique on two action recognition datasets demonstrates the ourperformance of our methodology compared to the state-of-the-art methods. Remarkably, we obtained 89 % accuracy on the well-known MSRAction3D dataset using only 3D trajectories of body joints obtained by Kinect

Abstract: We present a vision-inertial system which combines two RGB-Depth devices together with a wearable inertial movement unit in order to detect activities of the daily living. From multi-view videos, we extract dense trajectories enriched with a histogram of normals description computed from the depth cue and bag them into multi-view codebooks. During the later classification step a multi-class support vector machine with a RBF- 2 kernel combines the descriptions at kernel level. In order to perform action detection from the videos, a sliding window approach is utilized. On the other hand, we extract accelerations, rotation angles, and jerk features from the inertial data collected by the wearable placed on the user’s dominant wrist. During gesture spotting, a dynamic time warping is applied and the aligning costs to a set of pre-selected gesture sub-classes are thresholded to determine possible detections. The outputs of the two modules are combined in a late-fusion fashion. The system is validated in a real-case scenario with elderly from an elder home. Learning-based fusion results improve the ones from the single modalities, demonstrating the success of such multimodal approach.

Abstract: This paper describes a novel framework for detection and suppression of properly shadowed regions for most possible scenarios occurring in real video sequences. Our approach requires no prior knowledge about the scene, nor is it restricted to specific scene structures. Furthermore, the technique can detect both achromatic and chromatic shadows even in the presence of camouflage that occurs when foreground regions are very similar in color to shadowed regions. The method exploits local color constancy properties due to reflectance suppression over shadowed regions. To detect shadowed regions in a scene, the values of the background image are divided by values of the current frame in the RGB color space. We show how this luminance ratio can be used to identify segments with low gradient constancy, which in turn distinguish shadows from foreground. Experimental results on a collection of publicly available datasets illustrate the superior performance of our method compared with the most sophisticated, state-of-the-art shadow detection algorithms. These results show that our approach is robust and accurate over a broad range of shadow types and challenging video conditions.

Abstract: Impact factor 2010: 5.308
Impact factor 2011/12?: 5.96
Locating the center of the eyes allows for valuable information to be captured and used in a wide range of applications. Accurate eye center location can be determined using commercial eye-gaze trackers, but additional constraints and expensive hardware make these existing solutions unattractive and impossible to use on standard (i.e., visible wavelength), low-resolution images of eyes. Systems based solely on appearance are proposed in the literature, but their accuracy does not allow us to accurately locate and distinguish eye centers movements in these low-resolution settings. Our aim is to bridge this gap by locating the center of the eye within the area of the pupil on low-resolution images taken from a webcam or a similar device. The proposed method makes use of isophote properties to gain invariance to linear lighting changes (contrast and brightness), to achieve in-plane rotational invariance, and to keep low-computational costs. To further gain scale invariance, the approach is applied to a scale space pyramid. In this paper, we extensively test our approach for its robustness to changes in illumination, head pose, scale, occlusion, and eye rotation. We demonstrate that our system can achieve a significant improvement in accuracy over state-of-the-art techniques for eye center location in standard low-resolution imagery.

Abstract: The Quantitative Coronary Angiography (QCA) is a methodology used to evaluate the arterial diseases and, in particular, the degree of stenosis. In this paper we propose AQCA, a fully automatic method for vessel segmentation based on graph cut theory. Vesselness, geodesic paths and a new multi-scale edgeness map are used to compute a globally optimal artery segmentation. We evaluate the method performance in a rigorous numerical way on two datasets. The method can detect an artery with precision 92.9 +/- 5% and sensitivity 94.2 +/- 6%. The average absolute distance error between detected and ground truth centerline is 1.13 +/- 0.11 pixels (about 0.27 +/- 0.025 mm) and the absolute relative error in the vessel caliber estimation is 2.93% with almost no bias. Moreover, the method can discriminate between arteries and catheter with an accuracy of 96.4%.

Abstract: Scene text detection and recognition is a crucial task in computer vision with numerous real-world applications. Transformer-based approaches are behind all current state-of-the-art models and have achieved excellent performance. However, the computational requirements of the transformer architecture makes training these methods slow and resource heavy. In this paper, we introduce a new token pruning strategy that significantly decreases training and inference times without sacrificing performance, striking a balance between accuracy and speed. We have applied this pruning technique to our own end-to-end transformer-based scene text understanding architecture. Our method uses a separate detection branch to guide the pruning of uninformative image features, which significantly reduces the number of tokens at the input of the transformer. Experimental results show how our network is able to obtain competitive results on multiple public benchmarks while running at significantly higher speeds.