In this paper, we present a new equalizer based on probabilistic neural networks (PNN) for visible light communication (VLC) systems. While traditional artificial neural networks (ANN) type equalizers have to identify the distribution of the received signal, in our method kernel density estimation (KDE) is exploited to obtain the probability density function (PDF) of the states of the received signal, which reduces the computational complexity, especially when severe inter symbol interference (ISI) exists and when the number of signal states in the observation space increases. Then the constructed PDF is used to find the optimal Bayes solution at the receiver. Moreover, the decision feedback (DF) mechanism is utilized for further performance enhancements. The simulated bit error rate (BER) performances of the presented PNN based equalizers are compared with linear equalizer (LE) and conventional decision feedback equalizer (DFE). Simulation results demonstrate that the PNN-based equalizer without and with DF outperform LE and conventional DFE, respectively, thanks to the employment of KDE.

In this paper, we propose a novel algorithm that infers the 3D layout of building facades from a single 2D image of an urban scene. Different from existing methods that only yield coarse orientation labels or qualitative block approximations, our algorithm quantitatively reconstructs building facades in 3D space using a set of planes mutually related by 3D geometric constraints. Each plane is characterized by a continuous orientation vector and a depth distribution. An optimal solution is reached through inter-planar interactions. Due to the quantitative and plane-based nature of our geometric reasoning, our model is more expressive and informative than existing approaches. Experiments show that our method compares competitively with the state of the art on both 2D and 3D measures, while yielding a richer interpretation of the 3D scene behind the image.

Objects in a real world image cannot have arbitrary appearance, sizes and locations due to geometric constraints in 3D space. Such a 3D geometric context plays an important role in resolving visual ambiguities and achieving coherent object detection. In this paper, we develop a RANSAC-CRF framework to detect objects that are geometrically coherent in the 3D world. Different from existing methods, we propose a novel generalized RANSAC algorithm to generate global 3D geometry hypotheses from local entities such that outlier suppression and noise reduction is achieved simultaneously. In addition, we evaluate those hypotheses using a CRF which considers both the compatibility of individual objects under global 3D geometric context and the compatibility between adjacent objects under local 3D geometric context. Experiment results show that our approach compares favorably with the state of the art.

Cashiers in retail stores usually exhibit certain repetitive and periodic activities when processing items. Detecting such activities plays a key role in most retail fraud detection systems. In this paper, we propose a highly efficient, effective and robust vision technique to detect checkout-related primitive activities, based on a hierarchical finite state machine (FSM). Our deterministic approach uses visual features and prior spatial constraints on the hand motion to capture particular motion patterns performed in primitive activities. We also apply our approach to the problem of retail fraud detection. Experimental results on a large set of video data captured from retail stores show that our approach, while much simpler and faster, achieves significantly better results than state-of-the-art machine learning-based techniques both in detecting checkout-related activities and in detecting checkout-related fraudulent incidents.

We propose a robust abandoned object detection algorithm for real-time video surveillance. Different from conventional approaches that mostly rely on pixel-level processing, we perform region-level analysis in both background maintenance and static foreground object detection. In background maintenance, region-level information is fed back to adaptively control the learning rate. In static foreground object detection, region-level analysis double-checks the validity of candidate abandoned blobs. Attributed to such analysis, our algorithm is robust against illumination change, "ghosts" left by removed objects, distractions from partially static objects, and occlusions. Experiments on nearly 130,000 frames of i-LIDS dataset show the superior performance of our approach.

We propose a novel approach for enhancing precision in a leading video analytics system that detects cashier fraud in grocery stores for loss prevention. While intelligent video analytics has recently become a promising means of loss prevention for retailers, most of the real-world systems suffer from a large number of false alarms, resulting in a significant waste of human labor during manual verification. Our proposed approach starts with the candidate fraudulent events detected by a state-of-the-art system. Such fraudulent events are a set of visually recognized checkout-related activities of the cashier without barcode associations. Instead of conducting costly video analysis, we extract a few keyframes to represent the essence of each candidate fraudulent event, and compare those keyframes to identify whether or not the event is a valid check-out process that involves consistent appearance changes on the lead-in belt, the scan area and the take-away belt. Our approach also performs a margin-based soft classification so that the user could trade off between saving human labor and preserving high recall. Experiments on days of surveillance videos collected from real grocery stores show that our algorithm can save about 50% of human labor while preserving over 90% of true alarms with small computational overhead.

Detecting and segmenting cell regions in microscopic images is a challenging task, because cells typically do not have rich features, and their shapes and appearances are highly irregular and flexible. Furthermore, cells often form clusters, rendering the existing joint detection and segmentation algorithms unable to segment out individual cells. We address these difficulties by proposing a Heterogeneous Conditional Random Field (HCRF), in which different nodes have different state sets. The state sets are designed in such a way that the resulting HCRF model could encode all possible detection/segmentation cases while keeping its identifiability and compactness. Attributed to the provably optimal design of the state sets, the proposed model successfully realizes joint detection and segmentation of the cell regions into individual cells whether the cells are separate or touch one another. Experiments on two different types of cell images show that the HCRF outperforms several recently proposed methods.

We propose a complete solution to robust and accurate object tracking in face of various types of occlusions which pose many challenges to correct judgment of occlusion situation and proper update of target template. In order to tackle those challenges, we first propose a content-adaptive progressive occlusion analysis (CAPOA) algorithm. By combining the information provided by spatiotemporal context, reference target, and motion constraints together, the algorithm makes a clear distinction between the target and outliers. Accurate tracking of an occluded target is achieved by rectifying the target location using the variant-mask template matching (VMTM). In order to deal with template drift in the process of template update, we propose a drift-inhibitive masked Kalman appearance filter (DIMKAF) which accurately evaluates the influence of template drift when updating the masked template. Finally, we devise a local best match authentication (LBMA) algorithm to handle complete occlusions, so that we can achieve a much more trustworthy detection of the end of an arbitrarily long complete occlusion. Experimental results show that our proposed solution tracks targets reliably and accurately no matter when they are under: short-term, long-term, partial or complete occlusions.

In object tracking, occlusions significantly undermine the performance of tracking algorithms. Unlike the existing methods that solely depend on the observed target appearance to detect occluders, we propose an algorithm that progressively analyzes the occlusion situation by exploiting the spatiotemporal context information, which is further double checked by the reference target and motion constraints. This strategy enables our proposed algorithm to make a clearer distinction between the target and occluders than existing approaches. To further improve the tracking performance, we rectify the occlusion-interfered erroneous target location by employing a variant-mask template matching operation. As a result, correct target location can always be obtained regardless of the occlusion situation. Using these techniques, the robustness of tracking under occlusions is significantly promoted. Experimental results have confirmed the effectiveness of our proposed algorithm.

We propose a new method addressing the problem of template drift, a common phenomenon in which the target gradually shifts away from the template in object tracking. Much effort has been devoted to this problem, but the results are not satisfactory enough due to the lack of quantitative analysis of its cause. In this paper, after carefully examining where template drift stems from and how it influences template update, we derive expressions that accurately evaluate the model noises of the Kalman appearance filter employed to update the template. The appearance filter therefore achieves an optimal balance between reducing template drift and keeping track of target appearance variations. We perform experiments on a wide range of real-world video sequences containing diverse degrees of target appearance variations. All the experimental results confirm the effectiveness of our algorithm.