基于模糊逻辑的多相机非重叠场景的物体匹配

An approach based on fuzzy logic for matching both articulated and non-articulated objects across multiple non-overlapping field of views (FoVs) from multiple cameras is proposed. We call it fuzzy logic matching algorithm (FLMA). The approach uses the information of object motion, shape and camera topology for matching objects across camera views. The motion and shape information of targets are obtained by tracking them using a combination of ConDensation and CAMShift tracking algorithms. The information of camera topology is obtained and used by calculating the projective transformation of each view with the common ground plane. The algorithm is suitable for tracking non-rigid objects with both linear and non-linear motion. We show videos of tracking objects across multiple cameras based on FLMA. From our experiments, the system is able to correctly match the targets across views with a high accuracy.     

Correspondence-Free Activity Analysis and Scene Modeling in Multiple Camera Views

We propose a novel approach for activity analysis in multiple synchronized but uncalibrated static camera views. In this paper, we refer to activities as motion patterns of objects, which correspond to paths in far-field scenes. We assume that the topology of cameras is unknown and quite arbitrary, the fields of views covered by these cameras may have no overlap or any amount of overlap, and objects may move on different ground planes. Using low-level cues, objects are first tracked in each camera view independently, and the positions and velocities of objects along trajectories are computed as features. Under a probabilistic model, our approach jointly learns the distribution of an activity in the feature spaces of different camera views. Then, it accomplishes the following tasks: 1) grouping trajectories, which belong to the same activity but may be in different camera views, into one cluster; 2) modeling paths commonly taken by objects across multiple camera views; and 3) detecting abnormal activities. Advantages of this approach are that it does not require first solving the challenging correspondence problem, and that learning is unsupervised. Even though correspondence is not a prerequisite, after the models of activities have been learned, they can help to solve the correspondence problem, since if two trajectories in different camera views belong to the same activity, they are likely to correspond to the same object. Our approach is evaluated on a simulated data set and two very large real data sets, which have 22,951 and 14,985 trajectories, respectively.     

Key observation selection-based effective video synopsis for camera network

Nowadays, tremendous amount of video is captured endlessly from increased numbers of video cameras distributed around the world. Since needless information is abundant in the raw videos, making video browsing and retrieval is inefficient and time consuming. Video synopsis is an effective way to browse and index such video, by producing a short video representation, while keeping the essential activities of the original video. However, video synopsis for single camera is limited in its view scope, while understanding and monitoring overall activity for large scenarios is valuable and demanding. To solve the above issues, we propose a novel video synopsis algorithm for partially overlapping camera network. Our main contributions reside in three aspects: First, our algorithm can generate video synopsis for large scenarios, which can facilitate understanding overall activities. Second, for generating overall activity, we adopt a novel unsupervised graph matching algorithm to associate trajectories across cameras. Third, a novel multiple kernel similarity is adopted in selecting key observations for eliminating content redundancy in video synopsis. We have demonstrated the effectiveness of our approach on real surveillance videos captured by our camera network.     

Inference topology of distributed camera networks with multiple cameras

This paper proposes an inference method to construct the topology of a camera network with overlapping and non-overlapping fields of view for a commercial surveillance system equipped with multiple cameras. It provides autonomous object detection, tracking and recognition in indoor or outdoor urban environments. The camera network topology is estimated from object tracking results among and within FOVs. The merge-split method is used for object occlusion in a single camera and an EM-based approach for extracting the accurate object feature to track moving people and establishing object correspondence across multiple cameras. The appearance of moving people and the transition time between entry and exit zones is measured to track moving people across blind regions of multiple cameras with non-overlapping FOVs. Our proposed method graphically represents the camera network topology, as an undirected weighted graph using the transition probabilities and 8-directional chain code. The training phase and the test were run with eight cameras to evaluate the performance of our method. The temporal probability distribution and the undirected weighted graph are shown in the experiments.     

Spatial-temporal consistent labeling of tracked pedestrians across non-overlapping camera views

Tracking people across multiple cameras with non-overlapping views is a challenging task, since their observations are separated in time and space and their appearances may vary significantly. This paper proposes a Bayesian model to solve the consistent labeling problem across multiple non-overlapping camera views. Significantly different from related approaches, our model assumes neither people are well segmented nor their trajectories across camera views are estimated. We formulate a spatial-temporal probabilistic model in the hypothesis space that consists the potentially matched objects between the exit field of view (FOV) of one camera and the entry FOV of another camera. A competitive major color spectrum histogram representation (CMCSHR) for appearance matching between two objects is also proposed. The proposed spatial-temporal and appearance models are unified by a maximum-a-posteriori (MAP) Bayesian model. Based on this Bayesian model, when a detected new object corresponds to a group hypothesis (more than one object), we further develop an online method for online correspondence update using optimal graph matching (OGM) algorithm. Experimental results on three different real scenarios validate the proposed Bayesian model approach and the CMCSHR method. The results also show that the proposed approach is able to address the occlusion problem/group problem, i.e. finding the corresponding individuals in another camera view for a group of people who walk together into the entry FOV of a camera.     

Vehicle matching in smart camera networks using image projection profiles at multiple instances

Tracking vehicles using a network of cameras with non-overlapping views is a challenging problem of great importance in traffic surveillance. One of the main challenges is accurate vehicle matching across the cameras. Even if the cameras have similar views on vehicles, vehicle matching remains a difficult task due to changes of their appearance between observations, and inaccurate detections and occlusions, which often occur in real scenarios. To be executed on smart cameras the matching has also to be efficient in terms of needed data and computations. To address these challenges we present a low complexity method for vehicle matching robust against appearance changes and inaccuracies in vehicle detection. We efficiently represent vehicle appearances using signature vectors composed of Radon transform like projections of the vehicle images and compare them in a coarse-to-fine fashion using a simple combination of 1-D correlations. To deal with appearance changes we include multiple observations in each vehicle appearance model. These observations are automatically collected along the vehicle trajectory. The proposed signature vectors can be calculated in low-complexity smart cameras, by a simple scan-line algorithm of the camera software itself, and transmitted to the other smart cameras or to the central server. Extensive experiments based on real traffic surveillance videos recorded in a tunnel validate our approach.     

Distributed wide-area multi-object tracking with non-overlapping camera views

We present a distributed system for wide-area multi-object tracking across disjoint camera views. Every camera in the system performs multi-object tracking, and keeps its own trackers and trajectories. The data from multiple features are exchanged between adjacent cameras for object matching. We employ a probabilistic Petri Net-based approach to account for the uncertainties of the vision algorithms (such as unreliable background subtraction, and tracking failure) and to incorporate the available domain knowledge. We combine appearance features of objects as well as the travel-time evidence for target matching and consistent labeling across disjoint camera views. 3D color histogram, histogram of oriented gradients, local binary patterns, object size and aspect ratio are used as the appearance features. The distribution of the travel time is modeled by a Gaussian mixture model. Multiple features are combined by the weights, which are assigned based on the reliability of the features. By incorporating the domain knowledge about the camera configurations and the information about the received packets from other cameras, certain transitions are fired in the probabilistic Petri net. The system is trained to learn different parameters of the matching process, and updated online. We first present wide-area tracking of vehicles, where we used three non-overlapping cameras. The first and the third cameras are approximately 150 m apart from each other with two intersections in the blind region. We also present an example of applying our method to a people-tracking scenario. The results show the success of the proposed method. A comparison between our work and related work is also presented.     

基于场景模型的多摄像机目标跟踪算法

在多个相机组成的视频监视系统中,当目标物移出某一相机的视野而进入下一个时,如何实现相机的交接,实现目标物的继续跟踪是监视系统中要解决的关键问题。针对该问题,提出了一种基于位置比较的多摄像机运动目标跟踪方法。为获得目标物的位置,建立多个相机与目标物世界坐标之间映射关系的场景模型,并根据目标物出现在不同相机之间的视野边界线上的瞬间时刻的位置来给出重叠视野的边界线。由此可对任意角度摆放的多个具有重叠视野的相机之间运行的目标物进行接力跟踪。该方法可以适应多个目标物同时进入场景的情况,实验结果表明,该方法具有较高的鲁棒性,能够满足视频跟踪的实时性要求。     

Pose estimation from multiple cameras based on Sylvester’s equation

In this paper, we introduce a method to estimate the object’s pose from multiple cameras. We focus on direct estimation of the 3D object pose from 2D image sequences. Scale-Invariant Feature Transform (SIFT) is used to extract corresponding feature points from adjacent images in the video sequence. We first demonstrate that centralized pose estimation from the collection of corresponding feature points in the 2D images from all cameras can be obtained as a solution to a generalized Sylvester’s equation. We subsequently derive a distributed solution to pose estimation from multiple cameras and show that it is equivalent to the solution of the centralized pose estimation based on Sylvester’s equation. Specifically, we rely on collaboration among the multiple cameras to provide an iterative refinement of the independent solution to pose estimation obtained for each camera based on Sylvester’s equation. The proposed approach to pose estimation from multiple cameras relies on all of the information available from all cameras to obtain an estimate at each camera even when the image features are not visible to some of the cameras. The resulting pose estimation technique is therefore robust to occlusion and sensor errors from specific camera views. Moreover, the proposed approach does not require matching feature points among images from different camera views nor does it demand reconstruction of 3D points. Furthermore, the computational complexity of the proposed solution grows linearly with the number of cameras. Finally, computer simulation experiments demonstrate the accuracy and speed of our approach to pose estimation from multiple cameras.     

Distributed interactive video arrays for event capture and enhanced situational awareness

Video surveillance activity has dramatically increased over the past few years. Earlier work dealt mostly with single stationary cameras, but the recent trend is toward active multicamera systems. Such systems offer several advantages over single camera systems - multiple overlapping views for obtaining 3D information and handling occlusions, multiple nonoverlapping cameras for covering wide areas, and active pan-tilt-zoom (PTZ) cameras for observing object details. To address these issues, we have developed a multicamera video surveillance approach, called distributed interactive video array. The DIVA framework provides multiple levels of semantically meaningful information ("situational" awareness) to match the needs of multiple remote observers. We have designed DIVA-based systems that can track and identify vehicles and people, monitor perimeters and bridges, and analyze activities. A new video surveillance approach employing a large-scale cluster of video sensors demonstrates the promise of multicamera arrays for homeland security.     

Single Sensor Compressive Light Field Video Camera

This paper presents a novel compressed sensing (CS) algorithm and camera design for light field video capture using a single sensor consumer camera module. Unlike microlens light field cameras which sacrifice spatial resolution to obtain angular information, our CS approach is designed for capturing light field videos with high angular, spatial, and temporal resolution. The compressive measurements required by CS are obtained using a random color-coded mask placed between the sensor and aperture planes. The convolution of the incoming light rays from different angles with the mask results in a single image on the sensor; hence, achieving a significant reduction on the required bandwidth for capturing light field videos. We propose to change the random pattern on the spectral mask between each consecutive frame in a video sequence and extracting spatio-angular-spectral-temporal 6D patches. Our CS reconstruction algorithm for light field videos recovers each frame while taking into account the neighboring frames to achieve significantly higher reconstruction quality with reduced temporal incoherencies, as compared with previous methods. Moreover, a thorough analysis of various sensing models for compressive light field video acquisition is conducted to highlight the advantages of our method. The results show a clear advantage of our method for monochrome sensors, as well as sensors with color filter arrays.     

Robust object matching for persistent tracking with heterogeneous features

This paper addresses the problem of matching vehicles across multiple sightings under variations in illumination and camera poses. Since multiple observations of a vehicle are separated in large temporal and/or spatial gaps, thus prohibiting the use of standard frame-to-frame data association, we employ features extracted over a sequence during one time interval as a vehicle fingerprint that is used to compute the likelihood that two or more sequence observations are from the same or different vehicles. Furthermore, since our domain is aerial video tracking, in order to deal with poor image quality and large resolution and quality variations, our approach employs robust alignment and match measures for different stages of vehicle matching. Most notably, we employ a heterogeneous collection of features such as lines, points, and regions in an integrated matching framework. Heterogeneous features are shown to be important. Line and point features provide accurate localization and are employed for robust alignment across disparate views. The challenges of change in pose, aspect, and appearances across two disparate observations are handled by combining a novel feature-based quasi-rigid alignment with flexible matching between two or more sequences. However, since lines and points are relatively sparse, they are not adequate to delineate the object and provide a comprehensive matching set that covers the complete object. Region features provide a high degree of coverage and are employed for continuous frames to provide a delineation of the vehicle region for subsequent generation of a match measure. Our approach reliably delineates objects by representing regions as robust blob features and matching multiple regions to multiple regions using Earth Mover's Distance (EMD). Extensive experimentation under a variety of real-world scenarios and over hundreds of thousands of Confirmatory Identification (CID) trails has demonstrated about 95 percent accuracy in vehicle reacquisition with both visible and Infrared (IR) imaging cameras.     

The invariant features-based target tracking across multiple cameras

Target tracking across lenses is a popular research topic for video surveillance recently. This paper presents a method of target tracking across lenses with overlap regions. First, the target detection and tracking are completed with a single camera. Second, in order to obtain the location-invariant feature of the same target in the images with various cameras, the camera calibration is completed based on a three-dimension (3D) model. After that, for all images via multiple cameras, the coordinates of the 3D model are unified. Finally, referring to the assumption of spatial and temporal consistency of the target location across multiple cameras, the association among detected objects for the same target with different cameras is established. And a feature pool is built which contains perspective and scale features. Thus the same target is continuously tracked across multiple lenses. At last, the performance of the proposed approach is compared with KSP and PABC and demonstrated with indoor and outdoor experiments.     

Systems and algorithms for autonomous and scalable crowd surveillance using robotic PTZ cameras assisted by a wide-angle camera

We report an autonomous surveillance system with multiple pan-tilt-zoom (PTZ) cameras assisted by a fixed wide-angle camera. The wide-angle camera provides large but low resolution coverage and detects and tracks all moving objects in the scene. Based on the output of the wide-angle camera, the system generates spatiotemporal observation requests for each moving object, which are candidates for close-up views using PTZ cameras. Due to the fact that there are usually much more objects than the number of PTZ cameras, the system first assigns a subset of the requests/objects to each PTZ camera. The PTZ cameras then select the parameter settings that best satisfy the assigned competing requests to provide high resolution views of the moving objects. We propose an approximation algorithm to solve the request assignment and the camera parameter selection problems in real time. The effectiveness of the proposed system is validated in both simulation and physical experiment. In comparison with an existing work using simulation, it shows that in heavy traffic scenarios, our algorithm increases the number of observed objects by over 210%.     

Frame-level temporal calibration of video sequences from unsynchronized cameras

This paper describes a method for temporally calibrating video sequences from unsynchronized cameras by image processing operations, and presents two search algorithms to match and align trajectories across different camera views. Existing multi-camera systems assume that input video sequences are synchronized either by genlock or by time stamp information and a centralized server. Yet, hardware-based synchronization increases installation cost. Hence, using image information is necessary to align frames from the cameras whose clocks are not synchronized. The system built for temporal calibration is composed of three modules: object tracking module, calibration data extraction module, and the search module. A robust and efficient search algorithm is introduced that recovers the frame offset by matching the trajectories in different views, and finding the most reliable match. Thanks to information obtained from multiple trajectories, this algorithm is robust to possible errors in background subtraction and location extraction. Moreover, the algorithm can handle very large frame offsets. A RANdom SAmple Consensus (RANSAC) based version of this search algorithm is also introduced. Results obtained with different video sequences are presented, which show the robustness of the algorithms in recovering various range of frame offsets for video sequences with varying levels of object activity.     

Surveillance camera scheduling: a virtual vision approach

We present a surveillance system, comprising wide field-of-view (FOV) passive cameras and pan/tilt/zoom (PTZ) active cameras, which automatically captures high-resolution videos of pedestrians as they move through a designated area. A wide-FOV static camera can track multiple pedestrians, while any PTZ active camera can capture high-quality videos of one pedestrian at a time. We formulate the multi-camera control strategy as an online scheduling problem and propose a solution that combines the information gathered by the wide-FOV cameras with weighted round-robin scheduling to guide the available PTZ cameras, such that each pedestrian is observed by at least one PTZ camera while in the designated area. A centerpiece of our work is the development and testing of experimental surveillance systems within a visually and behaviorally realistic virtual environment simulator. The simulator is valuable as our research would be more or less infeasible in the real world given the impediments to deploying and experimenting with appropriately complex camera sensor networks in large public spaces. In particular, we demonstrate our surveillance system in a virtual train station environment populated by autonomous, lifelike virtual pedestrians, wherein easily reconfigurable virtual cameras generate synthetic video feeds. The video streams emulate those generated by real surveillance cameras monitoring richly populated public spaces.A preliminary version of this paper appeared as [1].     

Principal axis-based correspondence between multiple cameras for people tracking

Visual surveillance using multiple cameras has attracted increasing interest in recent years. Correspondence between multiple cameras is one of the most important and basic problems which visual surveillance using multiple cameras brings. In this paper, we propose a simple and robust method, based on principal axes of people, to match people across multiple cameras. The correspondence likelihood reflecting the similarity of pairs of principal axes of people is constructed according to the relationship between "ground-points" of people detected in each camera view and the intersections of principal axes detected in different camera views and transformed to the same view. Our method has the following desirable properties; 1) camera calibration is not needed; 2) accurate motion detection and segmentation are less critical due to the robustness of the principal axis-based feature to noise; 3) based on the fused data derived from correspondence results, positions of people in each camera view can be accurately located even when the people are partially occluded in all views. The experimental results on several real video sequences from outdoor environments have demonstrated the effectiveness, efficiency, and robustness of our method.     

非重叠监控视频中基于颜色空间分布模型的行人匹配方法

针对非重叠视域监控系统中行人目标关联问题,提出了一种基于颜色空间分布模型的行人目标匹配方法。使用最邻近聚类方法对目标的像素进行聚类,得到目标的MC特征;建立目标的颜色空间分布模型,并进行特征转换;在建立的分布模型的基础上,计算目标相似度并进行匹配。实验结果表明,该算法能有效地去除目标检测与分割产生的边缘背景像素的影响,在对相同目标保持较高匹配率的情况下,能较好地处理由于目标颜色分布差异而造成的误匹配情况,同时对于摄像机视角的差异以及目标姿态的变化有较好的鲁棒性。