Andreas Zweng

Introducing a statistical behavior model into camera-based fall detection

Camera based fall detection represents a solution to the problem of people falling down and being not able to stand up on their own again. For elderly people who live alone, such a fall is a major risk. In this paper we present an approach for fall detection based on multiple cameras supported by a statistical behavior model. The model describes the spatio-temporal unexpectedness of objects in a scene and is used to verify a fall detected by a semantic driven fall detection. In our work a fall is detected using multiple cameras where each of the camera inputs results in a separate fall confidence. These confidences are then combined into an overall decision and verified with the help of the statistical behavior model. This paper describes the fall detection approach as well as the verification step and shows results on 73 video sequences.

Unexpected Human Behavior Recognition in Image Sequences Using Multiple Features

This paper presents a novel approach for unexpected behavior recognition in image sequences with attention to high density crowd scenes. Due to occlusions, object-tracking in such scenes is challenging and in cases of low resolution or poor image quality it is not robust enough to efficiently detect abnormal behavior. The wide variety of possible actions performed by humans and the problem of occlusions makes action recognition unsuitable for behavior recognition in high density crowd scenes. The novel approach, which is presented in this paper uses features based on motion information instead of detecting actions or events in order to detect abnormality. Experiments demonstrate the potentials of the approach.

Performance evaluation of an improved relational feature model for pedestrian detection

In this paper, we evaluate a new algorithm for pedestrian detection using a relational feature model (RFM) in combination with histogram similarity functions. For histogram comparison, we use the bhattacharyya distance, histogram intersection, histogram correlation and the chi-square χ2 histogram similarity function. Relational features using the HOG descriptor compute the similarity between histograms of the HOG descriptor. The features are computed for all combinations of extracted histograms from a feature detection algorithm. Our experiments show, that the information of spatial histogram similarities reduces the number of false positives while preserving true positive detections. The detection algorithm is done, using a multi-scale overlapping sliding window approach. In our experiments, we show results for different sizes of the cell size from the HOG descriptor due to the large size of the resulting relational feature vector as well as different results from the mentioned histogram similarity functions. Additionally, the results show the influence of the amount of positive example images and negative example images during training on the classification performance of our approach.

Improved Relational Feature Model for People Detection Using Histogram Similarity Functions

In this paper, we propose a new approach for people detection using a relational feature model (RFM) in combination with histogram similarity functions such as the bhattacharyya distance, histogram intersection, histogram correlation and the chi-square χ2 histogram similarity function. The relational features are computed for all combinations of extracted features from a feature detection algorithm such as the Histograms of Oriented Gradients (HOG) feature descriptor. Our experiments show, that the information of spatial histogram similarities reduces the number of false positives while preserving true positive detections. The detection algorithm is done, using a multi-scale overlapping sliding window approach. In our experiments, we show results for different sizes of the cell size from the HOG descriptor due to the large size of the resulting relational feature vector as well as different results from the mentioned histogram similarity functions. Additionally our results show, that in addition to less false positives, true positive responses in regions near people are much more accurate using the relational features compared to non-relational feature models.

Evaluation of histogram-based similarity functions for different color spaces

In this paper we evaluate similarity functions for histograms such as chi-square and Bhattacharyya distance for different color spaces such as RGB or L*a*b*. Our main contribution is to show the performance of these histogram-based similarity functions combined with several color spaces. The evaluation is done on image sequences of the PETS 2009 dataset, where a sequence of frames is used to compute the histograms of three different persons in the scene. One of the most popular applications where similarity functions can be used is tracking. Data association is done in multiple stages where the first stage is the computation of the similarity of objects between two consecutive frames. Our evaluation concentrates on this first stage, where we use histograms as data type to compare the objects with each other. In this paper we present a comprehensive evaluation on a dataset of segmented persons with all combinations of the used similarity functions and color spaces.

Introducing a Inter-frame Relational Feature Model for Pedestrian Detection

Pedestrian detection has been used with the help of various local features in still images such as histograms of oriented gradients (HOG), local binary patterns (LBP) and more recently, the histograms of optical flow (HOF). In order to improve the robustness of pedestrian detection, movement of people can be taken into the training process which has been done in the HOF descriptor. Optical flow is used to model the movement of a person and to detect actions in image sequences. For action recognition it is necessary to incorporate movement into models when using feature descriptors such as the HOF descriptor. In this paper we introduce a novel method to train and to detect human movement for pedestrian detection using relational gradient features within multiple consecutive frames. The goal of this descriptor is to detect pedestrians using multiple frames for moving cameras instead of static cameras. The relational features between consecutive frames help to robustly find pedestrians in image sequences due to a flexible detection algorithm. We demonstrate the robustness of the resulting feature model computed for a temporal time window of three frames. In our experiments we show the improvement regarding true positives as well as false positives using our inter-frame HOG (ifHOG) model compared to other feature descriptors.

A FLEXIBLE RELATIONAL FEATURE MODEL FOR FALL DETECTION

Vision based fall detection solutions support elderly who live alone in their homes. For people falling down and not being able to stand up on their own again, such a fall is a major risk. In this work, we show a person detection approach using a relational feature model using consumer depth cameras. We propose a flexible relational feature model (FRFM) for fall detection in combination with histogram similarity functions such as the bhattacharyya distance, histogram intersection, histogram correlation and the chi-square 2 histogram similarity function. FRFM is an extension of the relational feature model (RFM) with the advantage that it can be used for all rotations of a body. The extension is necessary for fall detection due to the fact, that a lying person is rotated in the image where the standard person detection approach detects upright standing persons only. The relational features are computed for verification situations during fall detection on the basis of the Histograms of Oriented Gradients (HOG) feature descriptor. The experimental results show the best setup parameters for our feature model with different types of images (RGB and depth) and results on the specific field of fall detection.

Introducing confidence maps to increase the performance of person detectors

This paper deals with the problem of computational performance of person detection using the histogram of oriented gradients feature (HOG). Our approach increases the performance for implementations of person detection using a sliding window by learning the relationship of sizes of search windows and the position within the input image. In an offline training stage, confidence maps are computed at each scale of the search window and analyzed for a reduction of the number of used scales in the detection stage. Confidence maps are also computed during detection in order to make the classification more robust and to further increase the computational performance of the algorithm. Our approach shows a significant improvement of computational performance, while using only one core of the CPU and without using a graphics card in order to allow a low-cost solution of person detection using a sliding window approach.

High Performance Implementation of License Plate Recognition in Image Sequences

License plate recognition is done by recognizing the plate in single pictures. The license plate is analyzed in three steps namely the localization of the plate, the segmentation of the characters and the classification of the characters. Temporal redundant information has allready been used to improve the recognition rate, therefore fast algorithms have to be provided to get as many temporal classifications of a moving car as possible. In this paper a fast implementation for single classifications of license plates and performance increasing algorithms for statistical analysis other than a simple majority voting in image sequences are presented. The motivation of using the redundant information in image sequences and therefore classify one car multiple times is to have a more robust and converging classification where wrong single classifications can be suppressed.