Ralf Dragon

Shadow detection for moving humans using gradient-based background subtraction

Cast shadows cause serious problems in the functionality of vision-based applications, such as video surveillance, traffic monitoring and various other applications. Accurate detection and removal of cast shadows is a challenging task. Common shadow detection techniques normally use color information, which is not a reliable base in every scenario. This paper presents a novel scheme for real time detection of cast shadows using contour like structures of objects, which are obtained by gradient-based background subtraction. The scheme does not use any color information. Two basic rules are followed for shadow detection. The first rule is that shadows do not change the texture of the background. The second rule is a cast shadow lies outside the boundary of an object and has a relatively small common boundary with the object. Experimental results show the performance of the proposed scheme. Objective evaluation shows that the algorithm classifies 90 percent of the pixels of the objects and their shadow correctly.

View-invariant Fall Detection for Elderly in Real Home Environment

We propose a novel context based human fall detection mechanism in real home environment. Fall incidents are detected using head and floor information. The centroid location of the head and feet from each frame are used to learn a context model consisting of normal head and floor blocks. Every floor block has its associated Gaussian distribution, representing a set of head blocks. This Gaussian distribution defines standard vertical distance as average height of an object at that specific floor block. The classification of blocks and average height is later used to detect a fall. Fall detection methods often detect bending situations as fall. This method is able to distinguish bending and sitting from falling. Furthermore, a fall into any direction and at any distance from camera can be detected. Evaluation results show the robustness and high accuracy of the proposed approach.

Multi-scale clustering of frame-to-frame correspondences for motion segmentation

We present an approach for motion segmentation using independently detected keypoints instead of commonly used tracklets or trajectories. This allows us to establish correspondences over non- consecutive frames, thus we are able to handle multiple object occlusions consistently. On a frame-to-frame level, we extend the classical split-and-merge algorithm for fast and precise motion segmentation. Globally, we cluster multiple of these segmentations of different time scales with an accurate estimation of the number of motions. On the standard benchmarks, our approach performs best in comparison to all algorithms which are able to handle unconstrained missing data. We further show that it works on benchmark data with more than 98% of the input data missing. Finally, the performance is evaluated on a mobile-phone-recorded sequence with multiple objects occluded at the same time.

Altcare: Safe living for elderly people

Elderly people are the most growing segment of the population. Most of the elderly people like to live an independent life at their own homes in a familiar environment. Living alone is not always safe, and often emergency situations like a fall and unconsciousness occur. We present an ambient assistant living system Altcare, based on static network video cameras. The Altcare automatically learns the common events and their location. Furthermore it detects the emergency situations happening with elderly people. In case of some emergency situations, the Altcare first confirms it with the monitored person. In case of a positive response or no response at all, the emergency situation is notified to the responsible persons, in order to get timely and effective help. Along with personal information and a short medical history of the elderly person, a patch of masked video showing the emergency event is also transmitted. The masked video on one hand maintains the privacy of the elderly people and on the other hand it helps the administrators to diagnose the problem accurately. Masked video is also helpful to provide specific initial first aid. In order to ensure timely help, the system receives acknowledgement for the sent emergency call. Some of the elderly people desire to be checked on a periodic basis. Our system gives the administrators a facility to check the state of the person at any time, which gives the elderly person a sense of safety and mental satisfaction.

3D Object Recognition and Pose Estimation for Multiple Objects Using Multi-Prioritized RANSAC and Model Updating

We present a feature-based framework that combines spatial feature clustering, guided sampling for pose generation, and model updating for 3D object recognition and pose estimation. Existing methods fails in case of repeated patterns or multiple instances of the same object, as they rely only on feature discriminability for matching and on the estimator capabilities for outlier rejection. We propose to spatially separate the features before matching to create smaller clusters containing the object. Then, hypothesis generation is guided by exploiting cues collected off- and on-line, such as feature repeatability, 3D geometric constraints, and feature occurrence frequency. Finally, while previous methods overload the model with synthetic features for wide baseline matching, we claim that continuously updating the model representation is a lighter yet reliable strategy. The evaluation of our algorithm on challenging video sequences shows the improvement provided by our contribution.

Context-aware visual analysis of elderly activity in a cluttered home environment

This paper presents a semi-supervised methodology for automatic recognition and classification of elderly activity in a cluttered real home environment. The proposed mechanism recognizes elderly activities by using a semantic model of the scene under visual surveillance. We also illustrate the use of trajectory data for unsupervised learning of this scene context model. The model learning process does not involve any supervised feature selection and does not require any prior knowledge about the scene. The learned model in turn de-fines the activity and inactivity zones in the scene. An activity zone further contains block-level reference information, which is used to generate features for semi-supervised classification using transductive support vector machines. We used very few labeled examples for initial training. Knowledge of activity and inactivity zones improves the activity analysis process in realistic scenarios significantly. Experiments on real-life videos have validated our approach: we are able to achieve more than 90% accuracy for two diverse types of datasets.

Model based 3d segmentation and OCT image undistortion of percutaneous implants

Optical Coherence Tomography (OCT) is a noninvasive imaging technique which is used here for in vivo biocompatibility studies of percutaneous implants. A prerequisite for a morphometric analysis of the OCT images is the correction of optical distortions caused by the index of refraction in the tissue. We propose a fully automatic approach for 3D segmentation of percutaneous implants using Markov random fields. Refraction correction is done by using the subcutaneous implant base as a prior for model based estimation of the refractive index using a generalized Hough transform. Experiments show the competitiveness of our algorithm towards manual segmentations done by experts.

NF-features - no-feature-features for representing non-textured regions

In order to achieve a complete image description, we introduce no-feature-features (NF-features) representing object regions where regular interest point detectors do not detect features. As these regions are usually non-textured, stable re-localization in different images with conventional methods is not possible. Therefore, a technique is presented which re-localizes once-detected NF-features using correspondences of regular features. Furthermore, a distinctive NF descriptor for nontextured regions is derived which has invariance towards affine transformations and changes in illumination. For the matching of NF descriptors, an approach is introduced that is based on local image statistics. NF-features can be used complementary to all kinds of regular feature detection and description approaches that focus on textured regions, i.e. points, blobs or contours. Using SIFT, MSER, Hessian-Affine or SURF as regular detectors, we demonstrate that our approach is not only suitable for the description of non-textured areas but that precision and recall of the NF-features is significantly superior to those of regular features. In experiments with high variation of the perspective or image perturbation, at unchanged precision we achieve NF recall rates which are better by more than a factor of two compared to recall rates of regular features.

Fingerprints for machines: characterization and optical identification of grinding imprints

The profile of a 10mm wide and 1µm deep grinding imprint is as unique as a human fingerprint. To utilize this for fingerprinting mechanical components, a robust and strong characterization has to be used. We propose a feature-based approach, in which features of a 1D profile are detected and described in its 2D space-frequency representation. We show that the approach is robust on depth maps as well as intensity images of grinding imprints. To estimate the probability of misclassification, we derive a model and learn its parameters. With this model we demonstrate that our characterization has a false positive rate of approximately 10-20 which is as strong as a human fingerprint.

Towards feature-based situation assessment for airport apron video surveillance

We present a feature-based surveillance pipeline which, in contrast to traditional image-based methods, allows to learn a detailed description of the observed background as well as of foreground objects. The pipeline consists of motion segmentation of feature trajectories and subsequent tracking-by-recognition with updates. Furthermore, 3D object representations are learned in order to extract the 3D object pose of a later object recognition. Finally, we show how such sufficiently reliable information is inputted into a reasoning system comparing actual and nominal condition of an airport apron. By this, automatic situation assessment becomes possible in a manageable and reliable way.