A survey of human motion analysis using depth imagery

Analysis of human behaviour through visual information has been a highly active research topic in the computer vision community. This was previously achieved via images from a conventional camera, but recently depth sensors have made a new type of data available. This survey starts by explaining the advantages of depth imagery, then describes the new sensors that are available to obtain it. In particular, the Microsoft Kinect has made high-resolution real-time depth cheaply available. The main published research on the use of depth imagery for analysing human activity is reviewed. Much of the existing work focuses on body part detection and pose estimation. A growing research area addresses the recognition of human actions. The publicly available datasets that include depth imagery are listed, as are the software libraries that can acquire it from a sensor. This survey concludes by summarising the current state of work on this topic, and pointing out promising future research directions.

An study on ear detection and its applications to face detection

[EN]OpenCV includes di erent object detectors based on the Viola-Jones framework. Most of them are specialized to deal with the frontal face pattern and its inner elements: eyes, nose, and mouth. In this paper, we focus on the ear pattern detection, particularly when a head pro le or almost pro le view is present in the image. We aim at creating real-time ear detectors based on the general object detection framework provided with OpenCV. After training classi ers to detect left ears, right ears, and ears in general, the performance achieved is valid to be used to feed not only a head pose estimation system but also other applications such as those based on ear biometrics.

Combining face and facial feature detectors for face detection performance improvement

[EN]In this paper, we experimentally study the combination of face and facial feature detectors to improve face detection performance. The face detection problem, as suggeted by recent face detection challenges, is still not solved. Face detectors traditionally fail in large-scale problems and/or when the face is occluded or di erent head rotations are present. The combination of face and facial feature detectors is evaluated with a public database. The obtained results evidence an improvement in the positive detection rate while reducing the false detection rate. Additionally, we prove that the integration of facial feature detectors provides useful information for pose estimation and face alignment.

Stripe based clothes segmentation

[EN]In this paper, a clothes segmentation method for fashion parsing is described. This method does not rely in a previous pose estimation but people segmentation. Therefore, novel and classic segmentation techniques have been considered and improved in order to achieve accurate people segmentation. Unlike other methods described in the literature, the output is the bounding box and the predominant color of the different clothes and not a pixel level segmentation. The proposal is based on dividing the person area into an initial fixed number of stripes, that are later fused according to similar color distribution. To assess the quality of the proposed method the experiments are carried out with the Fashionista dataset that is widely used in the fashion parsing community.

Reconstruction from image correspondences

A single picture provides a largely incomplete representation of the scene one is looking at. Usually it reproduces only a limited spatial portion of the scene according to the standpoint and the viewing angle, besides it contains only instantaneous information. Thus very little can be understood on the geometrical structure of the scene, the position and orientation of the observer with respect to it remaining also hard to guess. When multiple views, taken from different positions in space and time, observe the same scene, then a much deeper knowledge is potentially achievable. Understanding inter-views relations enables construction of a collective representation by fusing the information contained in every single image. Visual reconstruction methods confront with the formidable, and still unanswered, challenge of delivering a comprehensive representation of structure, motion and appearance of a scene from visual information. Multi-view visual reconstruction deals with the inference of relations among multiple views and the exploitation of revealed connections to attain the best possible representation. This thesis investigates novel methods and applications in the field of visual reconstruction from multiple views. Three main threads of research have been pursued: dense geometric reconstruction, camera pose reconstruction, sparse geometric reconstruction of deformable surfaces. Dense geometric reconstruction aims at delivering the appearance of a scene at every single point. The construction of a large panoramic image from a set of traditional pictures has been extensively studied in the context of image mosaicing techniques. An original algorithm for sequential registration suitable for real-time applications has been conceived. The integration of the algorithm into a visual surveillance system has lead to robust and efficient motion detection with Pan-Tilt-Zoom cameras. Moreover, an evaluation methodology for quantitatively assessing and comparing image mosaicing algorithms has been devised and made available to the community. Camera pose reconstruction deals with the recovery of the camera trajectory across an image sequence. A novel mosaic-based pose reconstruction algorithm has been conceived that exploit image-mosaics and traditional pose estimation algorithms to deliver more accurate estimates. An innovative markerless vision-based human-machine interface has also been proposed, so as to allow a user to interact with a gaming applications by moving a hand held consumer grade camera in unstructured environments. Finally, sparse geometric reconstruction refers to the computation of the coarse geometry of an object at few preset points. In this thesis, an innovative shape reconstruction algorithm for deformable objects has been designed. A cooperation with the Solar Impulse project allowed to deploy the algorithm in a very challenging real-world scenario, i.e. the accurate measurements of airplane wings deformations.

Algorithms for on-line image registration from multiple views

Images of a scene, static or dynamic, are generally acquired at different epochs from different viewpoints. They potentially gather information about the whole scene and its relative motion with respect to the acquisition device. Data from different (in the spatial or temporal domain) visual sources can be fused together to provide a unique consistent representation of the whole scene, even recovering the third dimension, permitting a more complete understanding of the scene content. Moreover, the pose of the acquisition device can be achieved by estimating the relative motion parameters linking different views, thus providing localization information for automatic guidance purposes. Image registration is based on the use of pattern recognition techniques to match among corresponding parts of different views of the acquired scene. Depending on hypotheses or prior information about the sensor model, the motion model and/or the scene model, this information can be used to estimate global or local geometrical mapping functions between different images or different parts of them. These mapping functions contain relative motion parameters between the scene and the sensor(s) and can be used to integrate accordingly informations coming from the different sources to build a wider or even augmented representation of the scene. Accordingly, for their scene reconstruction and pose estimation capabilities, nowadays image registration techniques from multiple views are increasingly stirring up the interest of the scientific and industrial community. Depending on the applicative domain, accuracy, robustness, and computational payload of the algorithms represent important issues to be addressed and generally a trade-off among them has to be reached. Moreover, on-line performance is desirable in order to guarantee the direct interaction of the vision device with human actors or control systems. This thesis follows a general research approach to cope with these issues, almost independently from the scene content, under the constraint of rigid motions. This approach has been motivated by the portability to very different domains as a very desirable property to achieve. A general image registration approach suitable for on-line applications has been devised and assessed through two challenging case studies in different applicative domains. The first case study regards scene reconstruction through on-line mosaicing of optical microscopy cell images acquired with non automated equipment, while moving manually the microscope holder. By registering the images the field of view of the microscope can be widened, preserving the resolution while reconstructing the whole cell culture and permitting the microscopist to interactively explore the cell culture. In the second case study, the registration of terrestrial satellite images acquired by a camera integral with the satellite is utilized to estimate its three-dimensional orientation from visual data, for automatic guidance purposes. Critical aspects of these applications are emphasized and the choices adopted are motivated accordingly. Results are discussed in view of promising future developments.

The reconstruction of skeletal movement: the soft tissue artefact issue

In 3D human movement analysis performed using stereophotogrammetric systems and skin markers, bone pose can only be estimated in an indirect fashion. During a movement, soft tissue deformations make the markers move with respect to the underlying bone generating soft tissue artefact (STA). STA has devastating effects on bone pose estimation and its compensation remains an open question. The aim of this PhD thesis was to contribute to the solution of this crucial issue. Modelling STA using measurable trial-specific variables is a fundamental prerequisite for its removal from marker trajectories. Two STA model architectures are proposed. Initially, a thigh marker-level artefact model is presented. STA was modelled as a linear combination of joint angles involved in the movement. This model was calibrated using ex-vivo and in-vivo STA invasive measures. The considerable number of model parameters led to defining STA approximations. Three definitions were proposed to represent STA as a series of modes: individual marker displacements, marker-cluster geometrical transformations (MCGT), and skin envelope shape variations. Modes were selected using two criteria: one based on modal energy and another on the selection of modes chosen a priori. The MCGT allows to select either rigid or non-rigid STA components. It was also empirically demonstrated that only the rigid component affects joint kinematics, regardless of the non-rigid amplitude. Therefore, a model of thigh and shank STA rigid component at cluster-level was then defined. An acceptable trade-off between STA compensation effectiveness and number of parameters can be obtained, improving joint kinematics accuracy. The obtained results lead to two main potential applications: the proposed models can generate realistic STAs for simulation purposes to compare different skeletal kinematics estimators; and, more importantly, focusing only on the STA rigid component, the model attains a satisfactory STA reconstruction with less parameters, facilitating its incorporation in an pose estimator.

Automated detection of fiducial screws from CT/DVT volume data for image-guided ENT surgery

This paper presents an automated solution for precise detection of fiducial screws from three-dimensional (3D) Computerized Tomography (CT)/Digital Volume Tomography (DVT) data for image-guided ENT surgery. Unlike previously published solutions, we regard the detection of the fiducial screws from the CT/DVT volume data as a pose estimation problem. We thus developed a model-based solution. Starting from a user-supplied initialization, our solution detects the fiducial screws by iteratively matching a computer aided design (CAD) model of the fiducial screw to features extracted from the CT/DVT data. We validated our solution on one conventional CT dataset and on five DVT volume datasets, resulting in a total detection of 24 fiducial screws. Our experimental results indicate that the proposed solution achieves much higher reproducibility and precision than the manual detection. Further comparison shows that the proposed solution produces better results on the DVT dataset than on the conventional CT dataset.

Unifying energy minimization and mutual information maximization for robust 2D/3D registration of X-ray and CT images

Similarity measure is one of the main factors that affect the accuracy of intensity-based 2D/3D registration of X-ray fluoroscopy to CT images. Information theory has been used to derive similarity measure for image registration leading to the introduction of mutual information, an accurate similarity measure for multi-modal and mono-modal image registration tasks. However, it is known that the standard mutual information measure only takes intensity values into account without considering spatial information and its robustness is questionable. Previous attempt to incorporate spatial information into mutual information either requires computing the entropy of higher dimensional probability distributions, or is not robust to outliers. In this paper, we show how to incorporate spatial information into mutual information without suffering from these problems. Using a variational approximation derived from the Kullback-Leibler bound, spatial information can be effectively incorporated into mutual information via energy minimization. The resulting similarity measure has a least-squares form and can be effectively minimized by a multi-resolution Levenberg-Marquardt optimizer. Experimental results are presented on datasets of two applications: (a) intra-operative patient pose estimation from a few (e.g. 2) calibrated fluoroscopic images, and (b) post-operative cup alignment estimation from single X-ray radiograph with gonadal shielding.

Tracking of industrial objects by using CAD models

In this paper we present a model-based approach for real-time camera pose estimation in industrial scenarios. The line model which is used for tracking is generated by rendering a polygonal model and extracting contours out of the rendered scene. By un-projecting a point on the contour with the depth value stored in the z-buffer, the 3D coordinates of the contour can be calculated. For establishing 2D/3D correspondences the 3D control points on the contour are projected into the image and a perpendicular search for gradient maxima for every point on the contour is performed. Multiple hypotheses of 2D image points corresponding to a 3D control point make the pose estimation robust against ambiguous edges in the image.

Real-time Human Motion Capture with Simple Marker Sets and Monocular Video

In this paper we present a hybrid method to track human motions in real-time. With simplified marker sets and monocular video input, the strength of both marker-based and marker-free motion capturing are utilized: A cumbersome marker calibration is avoided while the robustness of the marker-free tracking is enhanced by referencing the tracked marker positions. An improved inverse kinematics solver is employed for real-time pose estimation. A computer-visionbased approach is applied to refine the pose estimation and reduce the ambiguity of the inverse kinematics solutions. We use this hybrid method to capture typical table tennis upper body movements in a real-time virtual reality application.

Registration and fusion of contrast-enhanced MRI myocardial substrate maps and X-ray angiograms

The aim of this work is to provide the necessary methods to register and fuse the endo-epicardial signal intensity (SI) maps extracted from contrast-enhanced magnetic resonance imaging (ceMRI) with X-ray coronary ngiograms using an intrinsic registrationbased algorithm to help pre-planning and guidance of catheterization procedures. Fusion of angiograms with SI maps was treated as a 2D-3D pose estimation, where each image point is projected to a Plücker line, and the screw representation for rigid motions is minimized using a gradient descent method. The resultant transformation is applied to the SI map that is then projected and fused on each angiogram. The proposed method was tested in clinical datasets from 6 patients with prior myocardial infarction. The registration procedure is optionally combined with an iterative closest point algorithm (ICP) that aligns the ventricular contours segmented from two ventriculograms.

A Vision-based Quadrotor Multi-robot Solution for the Indoor Autonomy Challenge of the 2013 International Micro Air Vehicle Competition

This paper presents a completely autonomous solution to participate in the Indoor Challenge of the 2013 International Micro Air Vehicle Competition (IMAV 2013). Our proposal is a multi-robot system with no centralized coordination whose robotic agents share their position estimates. The capability of each agent to navigate avoiding collisions is a consequence of the resulting emergent behavior. Each agent consists of a ground station running an instance of the proposed architecture that communicates over WiFi with an AR Drone 2.0 quadrotor. Visual markers are employed to sense and map obstacles and to improve the pose estimation based on Inertial Measurement Unit (IMU) and ground optical flow data. Based on our architecture, each robotic agent can navigate avoiding obstacles and other members of the multi-robot system. The solution is demonstrated and the achieved navigation performance is evaluated by means of experimental flights. This work also analyzes the capabilities of the presented solution in simulated flights of the IMAV 2013 Indoor Challenge. The performance of the CVG UPM team was awarded with the First Prize in the Indoor Autonomy Challenge of the IMAV 2013 competition.

Towards a real-time 3D object recognition pipeline on mobile GPGPU computing platforms using low-cost RGB-D sensors

In this project, we propose the implementation of a 3D object recognition system which will be optimized to operate under demanding time constraints. The system must be robust so that objects can be recognized properly in poor light conditions and cluttered scenes with significant levels of occlusion. An important requirement must be met: the system must exhibit a reasonable performance running on a low power consumption mobile GPU computing platform (NVIDIA Jetson TK1) so that it can be integrated in mobile robotics systems, ambient intelligence or ambient assisted living applications. The acquisition system is based on the use of color and depth (RGB-D) data streams provided by low-cost 3D sensors like Microsoft Kinect or PrimeSense Carmine. The range of algorithms and applications to be implemented and integrated will be quite broad, ranging from the acquisition, outlier removal or filtering of the input data and the segmentation or characterization of regions of interest in the scene to the very object recognition and pose estimation. Furthermore, in order to validate the proposed system, we will create a 3D object dataset. It will be composed by a set of 3D models, reconstructed from common household objects, as well as a handful of test scenes in which those objects appear. The scenes will be characterized by different levels of occlusion, diverse distances from the elements to the sensor and variations on the pose of the target objects. The creation of this dataset implies the additional development of 3D data acquisition and 3D object reconstruction applications. The resulting system has many possible applications, ranging from mobile robot navigation and semantic scene labeling to human-computer interaction (HCI) systems based on visual information.

Visual perception for the 3D recognition of geometric pieces in robotic manipulation

During grasping and intelligent robotic manipulation tasks, the camera position relative to the scene changes dramatically because the robot is moving to adapt its path and correctly grasp objects. This is because the camera is mounted at the robot effector. For this reason, in this type of environment, a visual recognition system must be implemented to recognize and “automatically and autonomously” obtain the positions of objects in the scene. Furthermore, in industrial environments, all objects that are manipulated by robots are made of the same material and cannot be differentiated by features such as texture or color. In this work, first, a study and analysis of 3D recognition descriptors has been completed for application in these environments. Second, a visual recognition system designed from specific distributed client-server architecture has been proposed to be applied in the recognition process of industrial objects without these appearance features. Our system has been implemented to overcome problems of recognition when the objects can only be recognized by geometric shape and the simplicity of shapes could create ambiguity. Finally, some real tests are performed and illustrated to verify the satisfactory performance of the proposed system.