Fast Image-Based Obstacle Detection From Unmanned Surface Vehicles.

Obstacle detection plays an important role in unmanned surface vehicles (USVs). The USVs operate in a highly diverse environments in which an obstacle may be a floating piece of wood, a scuba diver, a pier, or a part of a shoreline, which presents a significant challenge to continuous detection from images taken on board. This paper addresses the problem of online detection by constrained, unsupervised segmentation. To this end, a new graphical model is proposed that affords a fast and continuous obstacle image-map estimation from a single video stream captured on board a USV. The model accounts for the semantic structure of marine environment as observed from USV by imposing weak structural constraints. A Markov random field framework is adopted and a highly efficient algorithm for simultaneous optimization of model parameters and segmentation mask estimation is derived. Our approach does not require computationally intensive extraction of texture features and comfortably runs in real time. The algorithm is tested on a new, challenging, dataset for segmentation, and obstacle detection in marine environments, which is the largest annotated dataset of its kind. Results on this dataset show that our model outperforms the related approaches, while requiring a fraction of computational effort.

Enhancing indoor inertial pedestrian navigation using a shoe-worn marker.

We propose a novel hybrid inertial sensors-based indoor pedestrian dead reckoning system, aided by computer vision-derived position measurements. In contrast to prior vision-based or vision-aided solutions, where environmental markers were used-either deployed in known positions or extracted directly from it-we use a shoe-fixed marker, which serves as positional reference to an opposite shoe-mounted camera during foot swing, making our system self-contained. Position measurements can be therefore more reliably fed to a complementary unscented Kalman filter, enhancing the accuracy of the estimated travelled path for 78%, compared to using solely zero velocities as pseudo-measurements.

Tracking by identification using computer vision and radio.

We present a novel system for detection, localization and tracking of multiple people, which fuses a multi-view computer vision approach with a radio-based localization system. The proposed fusion combines the best of both worlds, excellent computer-vision-based localization, and strong identity information provided by the radio system, and is therefore able to perform tracking by identification, which makes it impervious to propagated identity switches. We present comprehensive methodology for evaluation of systems that perform person localization in world coordinate system and use it to evaluate the proposed system as well as its components. Experimental results on a challenging indoor dataset, which involves multiple people walking around a realistically cluttered room, confirm that proposed fusion of both systems significantly outperforms its individual components. Compared to the radio-based system, it achieves better localization results, while at the same time it successfully prevents propagation of identity switches that occur in pure computer-vision-based tracking.

A two-stage dynamic model for visual tracking.

We propose a new dynamic model which can be used within blob trackers to track the target's center of gravity. A strong point of the model is that it is designed to track a variety of motions which are usually encountered in applications such as pedestrian tracking, hand tracking, and sports. We call the dynamic model a two-stage dynamic model due to its particular structure, which is a composition of two models: a liberal model and a conservative model. The liberal model allows larger perturbations in the target's dynamics and is able to account for motions in between the random-walk dynamics and the nearly constant-velocity dynamics. On the other hand, the conservative model assumes smaller perturbations and is used to further constrain the liberal model to the target's current dynamics. We implement the two-stage dynamic model in a two-stage probabilistic tracker based on the particle filter and apply it to two separate examples of blob tracking: 1) tracking entire persons and 2) tracking of a person's hands. Experiments show that, in comparison to the widely used models, the proposed two-stage dynamic model allows tracking with smaller number of particles in the particle filter (e.g., 25 particles), while achieving smaller errors in the state estimation and a smaller failure rate. The results suggest that the improved performance comes from the model's ability to actively adapt to the target's motion during tracking.

A multistage registration method using texture features.

We present a novel, multistage registration method based on Laws' texture features. In general, a large number of texture features may be extracted from the original intensity images. For each of the texture features, a criterion function that measures the similarity between the images may be derived. The proposed registration method consists of two major steps. In the first step, a dataset of images with the corresponding gold standard is used. In this step, the selection and ranking of the texture features for registration is made. The selection and ranking of the features is based on their robustness, accuracy, and capture range. The selected features are then entered in the second step, where the actual registration is performed using a sequence of registration stages. Our method is based on the selection of the most robust feature for the first registration stage and the selection of accurate feature(s) for the subsequent stages. The texture features are daisy-chained so that the accuracy of the previous feature is sufficient for the capture range of the next feature. We tested our method on 11 2D image pairs containing digital reconstructed radiographs and electron portal imaging modalities, which were difficult to register using intensity features alone. With our method, we have successfully registered 75% of the initial displacements, ranging from 5 to 7.5 mm, with the target-registration error below 3 mm, whereas the traditional intensity-based approach delivered only 15% successfully registered cases.

Symmetric image registration.

This paper presents an original non-rigid image registration approach, which tends to improve the registration by establishing a symmetric image interdependence. In order to gather more information about the image transformation it measures the image similarity in both registration directions. The presented solution is based on the interaction between the images involved in the registration process. Images interact through forces, which according to Newton's action-reaction law form a symmetric relationship. These forces may transform both of the images, although in our implementation one of the images remains fixed. The experiments performed to demonstrate the advantages of the symmetric registration approach involve the registration of simple objects, the recovery of synthetic deformation, and the inter-patient registration of real images of the head. The results show that the symmetric approach improves both the registration consistency and the registration correctness.

Robust and efficient vision system for group of cooperating mobile robots with application to soccer robots.

In this paper a global vision scheme for estimation of positions and orientations of mobile robots is presented. It is applied to robot soccer application which is a fast dynamic game and therefore needs an efficient and robust vision system implemented. General applicability of the vision system can be found in other robot applications such as mobile transport robots in production, warehouses, attendant robots, fast vision tracking of targets of interest and entertainment robotics. Basic operation of the vision system is divided into two steps. In the first, the incoming image is scanned and pixels are classified into a finite number of classes. At the same time, a segmentation algorithm is used to find corresponding regions belonging to one of the classes. In the second step, all the regions are examined. Selection of the ones that are a part of the observed object is made by means of simple logic procedures. The novelty is focused on optimization of the processing time needed to finish the estimation of possible object positions. Better results of the vision system are achieved by implementing camera calibration and shading correction algorithm. The former corrects camera lens distortion, while the latter increases robustness to irregular illumination conditions.

Observation and analysis of large-scale human motion.

Many team sports include complex human movement, which can be observed at different levels of detail. Some aspects of the athlete's motion can be studied in detail using commercially available high-speed, high-accuracy biomechanical measurement systems. However, due to their limitations, these devices are not appropriate for studying large-scale motion during a game (for example, the motion of a player running across the entire playing field). We describe an alternative approach to studying such large-scale motion, and present a video-based, computer-aided system, developed specifically for the purpose of acquiring large-scale motion data. The baseline of our approach consists of sacrificing much of the spatial accuracy and temporal resolution of widely used biomechanical measurement systems, to obtain data on human movement that span large areas and long intervals of time. Data can be obtained for each of the observed athletes with reasonable amount of operator work. The system was developed using the recordings of a handball match. Several field tests were performed to assess measurement error, including comparison to one of the widely available biomechanical measurement systems. With the help of the system presented, we could obtain position data for all 14 handball players on a 40 x 20 m large court with RMS error better than 0.6 m, covering 1 h of action. Several results, obtained during the handball match study are presented, in order to highlight the importance of large-scale motion acquisition.