Search

Long--term video tracking is of great importance for many applications in real-world scenarios. A key component for achieving long--term tracking is the tracker's capability of updating its internal representation of targets (the appearance model) to changing conditions. Given the rapid but fragmented development of this research area, we propose a unified conceptual framework for appearance model adaptation that enables a principled comparison of different approaches. Moreover, we introduce a novel evaluation methodology that enables simultaneous analysis of tracking accuracy and tracking success, without the need of setting application-dependent thresholds. Based on the proposed framework and this novel evaluation methodology, we conduct an extensive experimental comparison of trackers that perform appearance model adaptation. Theoretical and experimental analyses allow us to identify the most effective approaches as well as to highlight design choices that favor resilience to errors during the update process. We conclude the paper with a list of key open research challenges that have been singled out by means of our experimental comparison.

This paper proposes an effective algorithm for recognizing objects and accurately estimating their 6DOF pose in scenes acquired by a RGB-D sensor. The proposed method is based on a combination of different recognition pipelines, each exploiting the data in a diverse manner and generating object hypotheses that are ultimately fused together in an Hypothesis Verification stage that globally enforces geometrical consistency between model hypotheses and the scene. Such a scheme boosts the overall recognition performance as it enhances the strength of the different recognition pipelines while diminishing the impact of their specific weaknesses. The proposed method outperforms the state-of-the-art on two challenging benchmark datasets for object recognition comprising 35 object models and, respectively, 176 and 353 scenes.

Pattern matching is widely used in signal processing, computer vision, and image and video processing. Full search equivalent algorithms accelerate the pattern matching process and, in the meantime, yield exactly the same result as the full search. This paper proposes an analysis and comparison of state-of-the-art algorithms for full search equivalent pattern matching. Our intention is that the data sets and tests used in our evaluation will be a benchmark for testing future pattern matching algorithms, and that the analysis concerning state-of-the-art algorithms could inspire new fast algorithms. We also propose extensions of the evaluated algorithms and show that they outperform the original formulations.

In the last decades several cost aggregation methods aimed at improving the robustness of stereo correspondence within local and global algorithms have been proposed. Given the recent developments and the lack of an appropriate comparison, in this paper we survey, classify and compare experimentally on a standard data set the main cost aggregation approaches proposed in literature. The experimental evaluation addresses both accuracy and computational requirements, so as to outline the best performing methods under these two criteria.

We present a method aimed at monitoring access to interlocks and secured entrance areas, which deploys two views in order to robustly perform intrusion detection and singularization. The main contributions are represented by an original approach to perform background subtraction, which is particularly robust against sudden illumination changes, shadows and photometric distortions, and by the use of a feature extraction and classification approach which allows to reliably determine an estimation of the number of people currently occupying the monitored area. Our system is designed to operate in very small interlocks and can work in a substantially unstructured environment.

This paper describes the mapping of a recently introduced template matching algorithm based on the Normalized Cross Correlation (NCC) on a general purpose processor endowed with SIMD (Single Instruction Multiple Data) multimedia instructions. The algorithm relies on the Bounded Partial Correlation (BPC) technique, which consists in deploying a sufficient condition to detect unsatisfactory matching candidates at a reduced computational cost. First, we briefly describe the BPC technique and highlight the related expensive computations. Then, based on the analysis of the major SIMD multimedia instruction set extensions available nowadays, we define a processor-independent multimedia instruction set and show how to carry out the most expensive BPC calculations using these pseudo-instructions. Finally, we provide experimental results obtained mapping the proposed algorithm on a mainstream multimedia SIMD instruction set (i.e. MMX). We compare these results with those obtained with the brute force NCC algorithm. The results show that the BPC technique is suited for a parallel SIMD-style mapping and that its effectiveness can be significantly improved using the multimedia instructions available nowadays in most general purpose CPUs.

This “Work In Progress” paper, reports on a novel approach to pedestrian tracking in confined areas which is based on a combined use of a stereo vision tracking system and a set of sensors available on a mobile platform. This platform is a hands-free, sensory augmented, wearable computer, designed to support visits to museums and to archaeological sites. The system is going to be demonstrated within the framework of a Cultural Heritage application at the “Musei Universitari di Palazzo Poggi” in Bologna”.

Nowadays, many visual surveillance systems exploit pan/tilt/zoom (PTZ) cameras to increase the field of view of a surveyed area. The background subtraction technique is widespread to detect moving objects with a high accuracy using one stationary camera. Extending such algorithms to work with moving cameras requires to have a background mosaic at one's disposal. Many solutions using mosaic background subtraction have been proposed, which offer real time capabilities or high quality of the detected objects. However, most of them rely on prior assumptions which limit the camera motion or the algorithm to work with a depth field of view only. In this work we propose some innovative solutions to achieve a real time mosaic background apt to work with existing background subtraction algorithms to yield excellent foreground object masks. Extensive experiments accomplished on challenging indoor and outdoor scenes permit to assess the quality of the mosaic as well as of the detected moving masks.

In this paper we present a new background generation algorithm and show experimental results aimed at assessing its performance comparatively with respect to two other representative approaches. The algorithm is able to extract a stationary background from a short bootstrap sequence in which moving objects can also be present. The method works with pixel-wise temporal statistics and consists of three subsequent stages. The first two try to isolate for each pixel the stationary background process from the possible foreground processes due to the moving objects covering the pixel, thus voting the background process temporal median as the good background value. The third stage completes the background generation by means of a non-parametric statistical model of the temporal camera noise, inferred from the statistics computed in the previous stage.

We present a novel approach to the change detection problem based on a coarse-to-fine strategy. The basic idea consists in assigning to an efficient preliminary coarse-level detection the task to filter out the well known possible false changes (e.g., those due to camera noise and small displacements, or to scene illumination changes). This provides the subsequent fine-level detection with reliable supermasks of the true changed areas in the scene. In this way, the fine-level detection can "focus the attention" on limited parts of the frames, thus yielding remarkable advantages in terms of computational efficiency. Here, just a coarse-level detection algorithm based on background subtraction and on the concept of structure is presented, to stress that any pixel-level algorithm can be used afterwards and benefit in terms of robustness as well as of computational efficiency.

We present a novel approach to change detection based on a coarse-to-fine strategy. An efficient coarse-level detection is proposed that filters out most of the possible false changes, thus attaining reliable and tight coarse-grain super-masks of the truly changed areas. The subsequent fine-level detection can thus "focus the attention" just on the "interesting" parts of the frame and perform a robust selective background updating procedure by considering the complement of these masks. Besides, the analysis of a strip of pixels surrounding each coarse-grain blob allows to infer information on light changes possibly occurring in the blob's vicinity. Although any algorithm can be used as the final fine-level detection, here we show how the approach applies to a particular algorithm we devised, based on a non-parametric statistical modelling of the camera noise.

In intelligent transportation systems (ITS's) vehicle tracking is necessary to permit high-level analysis, such as vehicle counting or classification. Nowadays, the need for a precise vehicle behavior analysis is growing mainly in urban intersections. Typical urban traffic scenes contain high-cluttered areas where static and dynamic occlusions take place and objects are missed. Tracking systems relying on monocular cameras are widespread. However, often they are misled by the complicated object interactions occurring in those areas, thus yielding errors in the higher level modules. The real-time tracking system we have conceived relies on an algorithm exploiting local and global information from corner points and whole object's features that allows us to keep track of many different objects in challenging urban scenarios. We assess our results through extensive on-field testing by manually extracting the ground truth from different sequences taken by real world traffic monitoring systems.

Previous work has shown that adaptive traffic smoothing can be successfully applied to a Shared Ethernet to enable it to support Statistical real-time communication at the Field level. This paper addresses the case for fuzzy traffic smoothing to realize a tradeoff between bounded delivery times for soft real-time packets and throughput for non real-time traffic over non-microsegmented Switched Ethernet networks. The paper proposes an optimization technique, based on the use of multiobjective optimization, to tune the parameters of the fuzzy controller and discusses the implementation of the optimized fuzzy smoother in a real scenario, also giving comparative performance assessments. Copyright © 2005 IFAC

The paper proposes a novel change-detection algorithm for automated video surveillance applications. The algorithm is based on the idea of incorporating into the background model a set of simple low-level features capable of capturing effectively "structural" (i.e. robust with respect to illumination variations) information. Thanks to this approach, and unlike most conventional change-detection algorithms, the proposed algorithm is capable of handling correctly still and slow objects as well as of working properly throughout very long time spans. Moreover, the algorithm can naturally interact with the higher-level processing modules found in advanced video-based surveillance systems in order to allow for flexible and intelligent background maintenance.

At the University of Bologna, an interdisciplinary project is active in order to investigate the possibility of implementing and testing a robotic system to help the mobility of visually impaired persons. The adopted solution is based on the conversion of stereo vision images into virtual "bas-relief" surfaces, perceivable by the user by means of a suitable (portable) haptic interface. In the paper, details on the adopted solutions and the developed prototypes are given as well as the experimental results obtained so far.

This paper presents a stereo algorithm that enables real time dense disparity measurements on standard personal computers. Unlike many other dense stereo algorithms, which are based on two matching phases, the proposed algorithm relies on a single matching phase and allows for rejecting unreliable matches by exploiting violations of the uniqueness constraint and analysing the behaviour of the correlation scores. The overall algorithm has been carefully optimised using very efficient calculation schemes and deploying massively the SIMD parallel processing capabilities available nowadays in state-of-the-art general purpose microprocessors. The paper describes the algorithm and the optimisation strategies, and provides experimental results obtained on stereo pairs with ground-truth as well as execution times measurements.

The paper describes an SIMD massively parallel computer conceived for robot vision, and presents an application where the system is installed on a mobile robot to support inspection and manipulation tasks. The system consists of a scalable array of up to 4 K single-bit processors controlled by a general-purpose microcomputer through a dedicated interface. The system's architecture is overviewed, addressing the available prototype as well as the functional units currently under design. The authors provide examples of image analysis goals that can be efficiently reached using spatially-organized parallel architectures such as SIMD arrays and give measurements of present performances. Using code profiles the authors also evaluate the speedup associated with the configuration being designed. >

The paper proposes a temporal filtering technique for the disparity measurements generated by area-based stereo-matching algorithms. The technique improves temporal consistency of disparity measurements by reducing the matching errors due to noise affecting the imaging system. Moreover, the technique is capable of increasing the number of correct matches by locating uncertain measurements with a criterion based on statistical assumptions that has proven to be more accurate and selective than those relying on texture operators only which are typically deployed with standard area-based stereo algorithms. © 2001 IEEE.

We propose a classification for a set of pixel-level algorithms employed in video surveillance applications and define a performance evaluation metric, based on an analysis of experimental data, for comparing the addressed algorithms. The results of such a comparison are presented and discussed. The set of algorithms considered in this work includes several algorithms widely known in the literature.

The VIDET project is aimed at investigating the possibility of developing a wearable robotic system for helping the mobility of visually impaired persons. The basic idea consists in the conversion of real-time depth data gathered through stereo-vision into a virtual, 'bas-relief' model perceivable by means of a haptic interface. In this paper we describe the real-time stereo system, review the basic principles of the two main haptic devices developed so far, and present new experimental results concerning extraction of depth data by the stereo system and haptic perception of the virtual model recovered from stereo-data.

Recent research activity on stereo matching has proved the efficacy of local approaches based on advanced cost aggregation strategies in accurately retrieving 3D information. However, accuracy is typically achieved at expense of computational efficiency, with best methods being far from meeting real-time requirements. On the other side, basic real-time local algorithms relying on a rectangular correlation window suffer from significant ambiguity along depth borders and untextured areas. This work proposes a novel local approach aimed at maximizing the speed-accuracy trade-off by means of an efficient segmentation-based cost aggregation strategy.

The ever-increasing speed of current general purpose processors, together with architectural enhancements such as multimedia-oriented instruction set extensions, allow for deploying standard PC-based systems in a number of com-putationally intensive computer vision tasks. This paper de-scribes the PC-based real-time stereo vision system devel-oped within the VIDET project, which is a research project aimed at the development of a mobility aid for the visu-ally impaired. VIDET's approach consists in the conversion of depth data gathered through a stereo vision system into a 3D model perceivable by the user by means of a wire-actuated haptic interface. The developed stereo matching algorithm makes massive use of recursion and multime-dia instructions to achieve the performance figures needed to sustain user's real-time interaction with the 3D model through the haptic interface.

Whole understanding of the surroundings is paramount to autonomous systems. Recent works have shown that deep neural networks can learn geometry (depth) and motion (optical flow) from a monocular video without any explicit supervision from ground truth annotations, particularly hard to source for these two tasks. In this paper, we take an additional step toward holistic scene understanding with monocular cameras by learning depth and motion alongside with semantics, with supervision for the latter provided by a pre-trained network distilling proxy ground truth images. We address the three tasks jointly by a) a novel training protocol based on knowledge distillation and self-supervision and b) a compact network architecture which enables efficient scene understanding on both power hungry GPUs and low-power embedded platforms. We thoroughly assess the performance of our framework and show that it yields state-of-the-art results for monocular depth estimation, optical flow and motion segmentation., Comment: CVPR 2020. Code will be available at https://github.com/CVLAB-Unibo/omeganet

Depth estimation from a single image represents a very exciting challenge in computer vision. While other image-based depth sensing techniques leverage on the geometry between different viewpoints (e.g., stereo or structure from motion), the lack of these cues within a single image renders ill-posed the monocular depth estimation task. For inference, state-of-the-art encoder-decoder architectures for monocular depth estimation rely on effective feature representations learned at training time. For unsupervised training of these models, geometry has been effectively exploited by suitable images warping losses computed from views acquired by a stereo rig or a moving camera. In this paper, we make a further step forward showing that learning semantic information from images enables to improve effectively monocular depth estimation as well. In particular, by leveraging on semantically labeled images together with unsupervised signals gained by geometry through an image warping loss, we propose a deep learning approach aimed at joint semantic segmentation and depth estimation. Our overall learning framework is semi-supervised, as we deploy groundtruth data only in the semantic domain. At training time, our network learns a common feature representation for both tasks and a novel cross-task loss function is proposed. The experimental findings show how, jointly tackling depth prediction and semantic segmentation, allows to improve depth estimation accuracy. In particular, on the KITTI dataset our network outperforms state-of-the-art methods for monocular depth estimation.

We introduce GFrames, a novel local reference frame (LRF) construction for 3D meshes and point clouds. GFrames are based on the computation of the intrinsic gradient of a scalar field defined on top of the input shape. The resulting tangent vector field defines a repeatable tangent direction of the local frame at each point; importantly, it directly inherits the properties and invariance classes of the underlying scalar function, making it remarkably robust under strong sampling artifacts, vertex noise, as well as non-rigid deformations. Existing local descriptors can directly benefit from our repeatable frames, as we showcase in a selection of 3D vision and shape analysis applications where we demonstrate state-of-the-art performance in a variety of challenging settings.

Deep convolutional neural networks trained end-to-end are the state-of-the-art methods to regress dense disparity maps from stereo pairs. These models, however, suffer from a notable decrease in accuracy when exposed to scenarios significantly different from the training set, e.g., real vs synthetic images, etc.). We argue that it is extremely unlikely to gather enough samples to achieve effective training/tuning in any target domain, thus making this setup impractical for many applications. Instead, we propose to perform unsupervised and continuous online adaptation of a deep stereo network, which allows for preserving its accuracy in any environment. However, this strategy is extremely computationally demanding and thus prevents real-time inference. We address this issue introducing a new lightweight, yet effective, deep stereo architecture, Modularly ADaptive Network (MADNet) and developing a Modular ADaptation (MAD) algorithm, which independently trains sub-portions of the network. By deploying MADNet together with MAD we introduce the first real-time self-adaptive deep stereo system enabling competitive performance on heterogeneous datasets., Comment: Accepted at CVPR2019 as oral presentation. Code Available https://github.com/CVLAB-Unibo/Real-time-self-adaptive-deep-stereo

Recent statistics on causes of aviation accidents and incidents demonstrate that to increase air-transportation safety, we must reduce human errors' impact on operations. So, the industry should first address human factors related to people in stressful roles to significantly minimize such errors. In particular, aviation maintenance employees work under high-pressure conditions- that is, they're under strict time constraints and must adhere to stringent guidelines. Because of such constraints, they might be prone to making errors. Unfortunately, many of these errors might not become apparent until an accident occurs. Although maintenance errors are a recognized threat to aviation safety, there are few simulation and computer-based tools for managing human factor issues in this field. The main advantages in using computer-based systems to train or support technicians are that computers don't forget and that they can help humans clearly understand facts. Such features can help reduce errors due to procedure violations, misinterpretation of facts, or insufficient training. Toward that end, augmented reality (AR) is a promising technology to build advanced interfaces using interactive and wearable visualization systems to implement new methods to display documentation as digital data and graphical databases. Nevertheless, many factors-such as cumbersome hardware, the need to put markers on the aircraft, and the need to quickly create digital content-seem to hinder its effective implementation in industry.

In the past few years detection of repeatable and distinctive keypoints on 3D surfaces has been the focus of intense research activity, due on the one hand to the increasing diffusion of low-cost 3D sensors, on the other to the growing importance of applications such as 3D shape retrieval and 3D object recognition. This work aims at contributing to the maturity of this field by a thorough evaluation of several recent 3D keypoint detectors. A categorization of existing methods in two classes, that allows for highlighting their common traits, is proposed, so as to abstract all algorithms to two general structures. Moreover, a comprehensive experimental evaluation is carried out in terms of repeatability, distinctiveness and computational efficiency, based on a vast data corpus characterized by nuisances such as noise, clutter, occlusions and viewpoint changes.

The literature on local invariant 3D features is growing, also fostered by the advent of cheap off-the-shelf 3D sensors. Although several recent proposals in the field include both a detector and a descriptor, some of the most successful and used descriptors do not define a companion detector. Moreover, as vouched by the related field of image features, detectors and descriptors defined within the same proposal do not necessarily yield the highest performance when used together. Hence, in this work we investigate on the effectiveness of the many possible combinations between state-of-the-art 3D detectors and descriptors, so as to identify optimal pairs as well as highlight well-matched detectors for those descriptors lacking a companion feature detection algorithm.

The paper investigates on canonical references used for local surface description and matching. We formulate a novel proposal and carry out an extensive experimental evaluation addressing two major surface matching scenarios, namely shape registration and object recognition. We provide also a methodological contribution as, unlike previous work in the field, we propose a repeatability metric that captures the actual impact of the adopted local reference frame algorithm within surface matching tasks based on local 3D descriptors. Our proposal outperforms existing algorithms by a wide margin on several datasets acquired with different devices, such as laser scanners, stereo cameras and the Kinect, and in experiments relying on randomly extracted feature as well as state-of-the art key points.

Visual search for mobile devices relies on transmitting wirelessly image descriptors to a remote server. Hence, descriptors needs to be compressed to reduce bandwidth occupancy and network latency. In this paper, we investigate on the compression ratio versus accuracy of two low complexity coding schemes, based on Arithmetic and Type coding, applied to SIFT local descriptors.

Visual search for mobile devices relies on transmitting wirelessly a compact representation of the query image, generally in the form of feature descriptors, to a remote server. Descriptors are therefore compressed, so as to reduce bandwidth occupancy and network latency. Given the impressive pace of growth of 3D video technology, we foresee 3D visual search applications for the mobile and the robotic market to become a reality. Accordingly, our work proposes a study on compressed 3D descriptors, a fundamental building block for such prospective applications. Based on analysis of several compression approaches, we develop and assess different schemes to achieve a compact version of a state-of-the-art 3D descriptor. Through experiments on a vast dataset we demonstrate the ability to achieve compression rates as high as 98% with a negligible loss in 3D visual search performance.

We propose a novel approach for verifying model hypotheses in cluttered and heavily occluded 3D scenes. Instead of verifying one hypothesis at a time, as done by most state-of-the-art 3D object recognition methods, we determine object and pose instances according to a global optimization stage based on a cost function which encompasses geometrical cues. Peculiar to our approach is the inherent ability to detect significantly occluded objects without increasing the amount of false positives, so that the operating point of the object recognition algorithm can nicely move toward a higher recall without sacrificing precision. Our approach outperforms state-of-the-art on a challenging dataset including 35 household models obtained with the Kinect sensor, as well as on the standard 3D object recognition benchmark dataset.

We propose a method to perform automatic segmentation of 3D scenes based on a standard classifier, whose learning model is continuously improved by means of new samples, and a grouping stage, that enforces local consistency among classified labels. The new samples are automatically delivered to the system by a feedback loop based on a feature selection approach that exploits the outcome of the grouping stage. By experimental results on several datasets we demonstrate that the proposed online learning paradigm is effective in increasing the accuracy of the whole 3D segmentation thanks to the improvement of the learning model of the classifier by means of newly acquired, unsupervised data.

Intense research activity on 3D data analysis tasks, such as object recognition and shape retrieval, has recently fostered the proposal of many techniques to perform detection of repeatable and distinctive key points in 3D surfaces. This high number of proposals has not been accompanied yet by a comprehensive comparative evaluation of the methods. Motivated by this, our work proposes a performance evaluation of the state-of-the-art in 3D key point detection, mainly addressing the task of 3D object recognition. The evaluation is carried out by analyzing the performance of several prominent methods in terms of robustness to noise (real and synthetic), presence of clutter, occlusions and point-of-view variations.

Object segmentation in 3D data such as 3D meshes and range maps is an emerging topic attracting increasing research interest. This work proposes a novel method to perform segmentation relying on the use of 3D features. The deployment of a specific grouping algorithm based on a Markov Random Field model successively to classification allows at the same time yielding automatic segmentation of 3D data as well as deploying non-linear classifiers that can well adapt to the data characteristics. Moreover, we embed our approach in a framework that jointly exploits shape and texture information to improve the outcome of the segmentation stage. In addition to quantitative results on several 3D and 2.5D scenes, we also demonstrate the effectiveness of our approach on an online framework based on a stereo sensor.

This work aims at evaluating stereo matching algorithms in a 3D object recognition scenario, wherein objects have to be found and their 3D pose estimated efficiently and in presence of clutter and occlusions. Unlike all other surveys and evaluations of stereo methods, which compare accuracy based on a dataset of disparity maps with ground-truth, this work proposes an evaluation in terms of recognition ability which leverages on state-of-the-art approaches for 3D object recognition. The proposed evaluation methodology comprises a novel dataset characterized by realistic working conditions and compares state-of-the-art stereo algorithms potentially suitable to 3D object recognition applications.

Template matching is a computationally intensive problem aimed at locating a template within a image. When dealing with images having more than one channel, the computational burden becomes even more dramatic. For this reason, in this paper we investigate on a methodology to speed-up template matching on multi-channel images without deteriorating the outcome of the search. In particular, we propose a fast, exhaustive technique based on the Zero-mean Normalized Cross-Correlation (ZNCC) inspired from previous work related to grayscale images. Experimental testing performed over thousands of template matching instances demonstrates the efficiency of our proposal.

Several recent proposals have shown the feasibility of significantly speeding-up pattern matching by means of Full Search-equivalent techniques, i.e. without approximating the outcome of the search with respect to a brute force investigation. These techniques are generally heavily based on efficient incremental calculation schemes aimed at avoiding unnecessary computations. In a very recent and extensive experimental evaluation, Low Resolution Pruning turned out to be in most cases the best performing approach. In this paper we propose a computational analysis of several incremental techniques specifically designed to enhance the efficiency of LRP. In addition, we propose a novel LRP algorithm aimed at minimizing the theoretical number of operations by adaptively exploiting different incremental approaches. We demonstrate the effectiveness of our proposal by means of experimental evaluation on a large dataset.

A major issue for video surveillance embedded systems is the need to continuously perform a number of highly demanding operations even when the analyzed scene does not show peculiar or interesting features, so that power consumption is a critical issue. In this paper we present a low-power multimodal embedded video surveillance system aimed at detecting objects abandoned/removed in/from a static monitored scene. Energy-awareness is achieved by means of an efficient and scalable objects abandon/removal detection algorithm, a Linux governor that controls CPU frequency and operating mode so as to establish an optimal trade-off between fulfilling the application efficiency-accuracy requirements and maximizing battery life and, finally, a pyroelectric infrared sensor that allows to wake up the CPU only when video processing is actually needed.

Motivated by the increasing availability of 3D sensors capable of delivering both shape and texture information, this paper presents a novel descriptor for feature matching in 3D data enriched with texture. The proposed approach stems from the theory of a recently proposed descriptor for 3D data which relies on shape only, and represents its generalization to the case of multiple cues associated with a 3D mesh. The proposed descriptor, dubbed CSHOT, is demonstrated to notably improve the accuracy of feature matching in challenging object recognition scenarios characterized by the presence of clutter and occlusions.

In this paper we investigate Bayesian visual tracking based on change detection. Although in many proposals change detection is key for tracking, little attention has been paid to sound modeling of the interaction between the change detector and the tracker. In this work, we develop a principled framework whereby both processes can virtuously influence each other according to a Bayesian loop: change detection provides a completely specified observation likelihood to the tracker and the tracker provides an informative prior to the change detector.

Despite the significant number of stereo vision algorithms proposed in literature in the last decade, most proposals are notably computationally demanding and/or memory hungry so that it is unfeasible to employ them in application scenarios requiring real-time or near real-time processing on platforms with limited resources such as embedded devices. In this paper, we have selected the subset of proposals that appears more suited to the above requirements and, since literature lacks a proper comparison between these methods, we propose a quantitative experimental evaluation aimed at highlighting the best performing approach under the two criteria of accuracy and efficiency. The evaluation is performed on a standard benchmark dataset as well as on a novel dataset, acquired by means of an active technique, characterized by realistic working conditions.

Mobile Visual Search is a new technology that aims at linking physical objects with digital information by taking pictures of them using a mobile device with on-board camera. In this paper we present an easily extendible and adaptable framework for Mobile Visual Search applications that takes advantage of the Smart-M3 interoperability platform. After introducing Smart-M3, we describe the proposed architecture and illustrate its use in two novel visual search scenarios.

The use of robust feature descriptors is now key for many 3D tasks such as 3D object recognition and surface alignment. Many descriptors have been proposed in literature which are based on a non-unique local Reference Frame and hence require the computation of multiple descriptions at each feature points. In this paper we show how to deploy a unique local Reference Frame to improve the accuracy and reduce the memory footprint of the well-known 3D Shape Context descriptor. We validate our proposal by means of an experimental analysis carried out on a large dataset of 3D scenes and addressing an object recognition scenario.

In this paper we propose a formalization of change detection as a Bayesian order-consistency test, based on the assumption that disturbance factors such as illumination changes and variations of camera parameters do not change the ordering between noiseless intensities within a neighborhood of pixels. The assumption of additive, zero-mean, i.i.d. gaussian noise allows for testing the composite order-consistency hypothesis by efficient computation of the marginal likelihood. Moreover, since the above formalization enables to incorporate changed/unchanged class priors seamlessly, we also propose a simple method to derive informative priors based on the calculation of marginal likelihoods at reduced resolution. Experimental results on challenging test sequences characterized by sudden and strong illumination changes prove the effectiveness of the proposed approach.

Low-cost and low-power video surveillance systems based on networks of wireless video sensors will enter soon the marketplace with the promise of flexibility, quick deployment and providing accurate and real-time visual data. Energy autonomy and efficiency of the implemented algorithms are undoubtedly the primary design challenges to be addressed on systems subject to low computational capabilities and memory constraints. In this paper we present a low-power video sensor node designed for low-cost video surveillance which is able to detect abandoned and removed objects. The system exploits multi-modal sensor integration which saves on-board power consumption. In particular a Pyroelectric InfraRed (PIR)sensor is exploited to optimize the use of the camera,grabbing images only when required in order to obtain the maximum efficiency from event recognition. Our fixed-point ARM-based approach is characterized in terms of runtime execution and power consumption, while efficiency is demonstrated by experimental results and compared with floating point implementations.

The ability to augment a video stream with consistent virtual contents is an attractive Computer Vision application. The first Augmented Reality (AR) proposals required the scene to be endowed with special markers. Recently, thanks to the developments in the field of natural invariant local features, similar results have been achieved in a markerless scenario. The computer vision community is now equipped with a set of relatively standard techniques to solve the underlying markerless camera pose estimation problem, at least for planar textured reference objects. The majority of proposals, however, does not exploit temporal consistency across frames in order to reduce some disturbing effects of per-frame estimation, namely visualization of short spurious estimations and jitter. We proposes a new method based on Support Vector Regression to mitigate these undesired effects while preserving the ability to work in real-time. Our proposal can be used as a post processing step independent of the chosen pose estimation method, thus providing an effective and easily integrable building block for AR applications.

The paper proposes a novel human machine interface for gaming applications based on computer vision. The key idea is to allow the user to interact with the game by simply moving a hand-held consumer grade camera. Detection of natural features in the incoming video stream avoids instrumenting the scene with optical markers while preserving real-time computation and accuracy. The paper presents also a prototype videogame developed as proof-of-concept of our camera-based gaming interface. Thanks to recent advances in real-time extraction and matching of natural features from images on mobile platforms, our proposal holds the potential to enable a new generation of camera-controlled videogames for hand-held mobile devices.

Several methods aimed at effectively speeding up the block matching and template matching tasks have been recently proposed. A class of these methods, referred to as exhaustive due to the fact that they optimally solve the minimization problem of the matching cost, often deploys a succession of bounding functions based on a partitioning of the template and subwindow to perform rapid and reliable detection of non-optimal candidates. In this paper we propose a study aimed at improving the efficiency of one of these methods, that is, a state-of-the-art template matching technique known as Incremental Dissimilarity Approximations (IDA). In particular, we outline a methodology to order the succession of bounding functions deployed by this technique based on the analysis of the template only. Experimental results prove that the proposed approach is able to achieve improved efficiency.