Automatic Audio Segmentation Using the Generalized Likelihood Ratio

This paper presents a novel technique for segmenting an audio stream into homogeneous regions according to speaker identities, background noise, music, environmental and channel conditions. Audio segmentation is useful in audio diarization systems, which aim to annotate an input audio stream with information that attributes temporal regions of the audio into their specific sources. The segmentation method introduced in this paper is performed using the Generalized Likelihood Ratio (GLR), computed between two adjacent sliding windows over preprocessed speech. This approach is inspired by the popular segmentation method proposed by the pioneering work of Chen and Gopalakrishnan, using the Bayesian Information Criterion (BIC) with an expanding search window. This paper will aim to identify and address the shortcomings associated with such an approach. The result obtained by the proposed segmentation strategy is evaluated on the 2002 Rich Transcription (RT-02) Evaluation dataset, and a miss rate of 19.47% and a false alarm rate of 16.94% is achieved at the optimal threshold.

SAIVT-QUT@TRECVid 2012 : interactive surveillance event detection

In this paper, we propose an approach which attempts to solve the problem of surveillance event detection, assuming that we know the definition of the events. To facilitate the discussion, we first define two concepts. The event of interest refers to the event that the user requests the system to detect; and the background activities are any other events in the video corpus. This is an unsolved problem due to many factors as listed below: 1) Occlusions and clustering: The surveillance scenes which are of significant interest at locations such as airports, railway stations, shopping centers are often crowded, where occlusions and clustering of people are frequently encountered. This significantly affects the feature extraction step, and for instance, trajectories generated by object tracking algorithms are usually not robust under such a situation. 2) The requirement for real time detection: The system should process the video fast enough in both of the feature extraction and the detection step to facilitate real time operation. 3) Massive size of the training data set: Suppose there is an event that lasts for 1 minute in a video with a frame rate of 25fps, the number of frames for this events is 60X25 = 1500. If we want to have a training data set with many positive instances of the event, the video is likely to be very large in size (i.e. hundreds of thousands of frames or more). How to handle such a large data set is a problem frequently encountered in this application. 4) Difficulty in separating the event of interest from background activities: The events of interest often co-exist with a set of background activities. Temporal groundtruth typically very ambiguous, as it does not distinguish the event of interest from a wide range of co-existing background activities. However, it is not practical to annotate the locations of the events in large amounts of video data. This problem becomes more serious in the detection of multi-agent interactions, since the location of these events can often not be constrained to within a bounding box. 5) Challenges in determining the temporal boundaries of the events: An event can occur at any arbitrary time with an arbitrary duration. The temporal segmentation of events is difficult and ambiguous, and also affected by other factors such as occlusions.

A multi-layered approach for site detection in UAS emergency landing scenarios using geometry-based image segmentation

This paper presents an alternative approach to image segmentation by using the spatial distribution of edge pixels as opposed to pixel intensities. The segmentation is achieved by a multi-layered approach and is intended to find suitable landing areas for an aircraft emergency landing. We combine standard techniques (edge detectors) with novel developed algorithms (line expansion and geometry test) to design an original segmentation algorithm. Our approach removes the dependency on environmental factors that traditionally influence lighting conditions, which in turn have negative impact on pixel-based segmentation techniques. We present test outcomes on realistic visual data collected from an aircraft, reporting on preliminary feedback about the performance of the detection. We demonstrate consistent performances over 97% detection rate.

Nonparametric semantic segmentation for 3D street scenes

In this paper we propose a method to generate a large scale and accurate dense 3D semantic map of street scenes. A dense 3D semantic model of the environment can significantly improve a number of robotic applications such as autonomous driving, navigation or localisation. Instead of using offline trained classifiers for semantic segmentation, our approach employs a data-driven, nonparametric method to parse scenes which easily scale to a large environment and generalise to different scenes. We use stereo image pairs collected from cameras mounted on a moving car to produce dense depth maps which are combined into a global 3D reconstruction using camera poses from stereo visual odometry. Simultaneously, 2D automatic semantic segmentation using a nonparametric scene parsing method is fused into the 3D model. Furthermore, the resultant 3D semantic model is improved with the consideration of moving objects in the scene. We demonstrate our method on the publicly available KITTI dataset and evaluate the performance against manually generated ground truth.

Segmentation of cone-beam CT using a hidden Markov random field with informative priors

Purpose: Flat-detector, cone-beam computed tomography (CBCT) has enormous potential to improve the accuracy of treatment delivery in image-guided radiotherapy (IGRT). To assist radiotherapists in interpreting these images, we use a Bayesian statistical model to label each voxel according to its tissue type. Methods: The rich sources of prior information in IGRT are incorporated into a hidden Markov random field (MRF) model of the 3D image lattice. Tissue densities in the reference CT scan are estimated using inverse regression and then rescaled to approximate the corresponding CBCT intensity values. The treatment planning contours are combined with published studies of physiological variability to produce a spatial prior distribution for changes in the size, shape and position of the tumour volume and organs at risk (OAR). The voxel labels are estimated using the iterated conditional modes (ICM) algorithm. Results: The accuracy of the method has been evaluated using 27 CBCT scans of an electron density phantom (CIRS, Inc. model 062). The mean voxel-wise misclassification rate was 6.2%, with Dice similarity coefficient of 0.73 for liver, muscle, breast and adipose tissue. Conclusions: By incorporating prior information, we are able to successfully segment CBCT images. This could be a viable approach for automated, online image analysis in radiotherapy.

Segmentation of cone-beam CT using a hidden Markov random field with informative priors

Cone-beam computed tomography (CBCT) has enormous potential to improve the accuracy of treatment delivery in image-guided radiotherapy (IGRT). To assist radiotherapists in interpreting these images, we use a Bayesian statistical model to label each voxel according to its tissue type. The rich sources of prior information in IGRT are incorporated into a hidden Markov random field model of the 3D image lattice. Tissue densities in the reference CT scan are estimated using inverse regression and then rescaled to approximate the corresponding CBCT intensity values. The treatment planning contours are combined with published studies of physiological variability to produce a spatial prior distribution for changes in the size, shape and position of the tumour volume and organs at risk. The voxel labels are estimated using iterated conditional modes. The accuracy of the method has been evaluated using 27 CBCT scans of an electron density phantom. The mean voxel-wise misclassification rate was 6.2\%, with Dice similarity coefficient of 0.73 for liver, muscle, breast and adipose tissue. By incorporating prior information, we are able to successfully segment CBCT images. This could be a viable approach for automated, online image analysis in radiotherapy.

Automatic segmentation of choroidal thickness in optical coherence tomography

The assessment of choroidal thickness from optical coherence tomography (OCT) images of the human choroid is an important clinical and research task, since it provides valuable information regarding the eye’s normal anatomy and physiology, and changes associated with various eye diseases and the development of refractive error. Due to the time consuming and subjective nature of manual image analysis, there is a need for the development of reliable objective automated methods of image segmentation to derive choroidal thickness measures. However, the detection of the two boundaries which delineate the choroid is a complicated and challenging task, in particular the detection of the outer choroidal boundary, due to a number of issues including: (i) the vascular ocular tissue is non-uniform and rich in non-homogeneous features, and (ii) the boundary can have a low contrast. In this paper, an automatic segmentation technique based on graph-search theory is presented to segment the inner choroidal boundary (ICB) and the outer choroidal boundary (OCB) to obtain the choroid thickness profile from OCT images. Before the segmentation, the B-scan is pre-processed to enhance the two boundaries of interest and to minimize the artifacts produced by surrounding features. The algorithm to detect the ICB is based on a simple edge filter and a directional weighted map penalty, while the algorithm to detect the OCB is based on OCT image enhancement and a dual brightness probability gradient. The method was tested on a large data set of images from a pediatric (1083 B-scans) and an adult (90 B-scans) population, which were previously manually segmented by an experienced observer. The results demonstrate the proposed method provides robust detection of the boundaries of interest and is a useful tool to extract clinical data.

Modeling the viability of the free-ranging cheetah population in Namibia : an object-oriented Bayesian network approach

Conservation of free-ranging cheetah (Acinonyx jubatus) populations is multi faceted and needs to be addressed from an ecological, biological and management perspective. There is a wealth of published research, each focusing on a particular aspect of cheetah conservation. Identifying the most important factors, making sense of various (and sometimes contrasting) findings, and taking decisions when little or no empirical data is available, are everyday challenges facing conservationists. Bayesian networks (BN) provide a statistical modeling framework that enables analysis and integration of information addressing different aspects of conservation. There has been an increased interest in the use of BNs to model conservation issues, however the development of more sophisticated BNs, utilizing object-oriented (OO) features, is still at the frontier of ecological research. We describe an integrated, parallel modeling process followed during a BN modeling workshop held in Namibia to combine expert knowledge and data about free-ranging cheetahs. The aim of the workshop was to obtain a more comprehensive view of the current viability of the free-ranging cheetah population in Namibia, and to predict the effect different scenarios may have on the future viability of this free-ranging cheetah population. Furthermore, a complementary aim was to identify influential parameters of the model to more effectively target those parameters having the greatest impact on population viability. The BN was developed by aggregating diverse perspectives from local and independent scientists, agents from the national ministry, conservation agency members and local fieldworkers. This integrated BN approach facilitates OO modeling in a multi-expert context which lends itself to a series of integrated, yet independent, subnetworks describing different scientific and management components. We created three subnetworks in parallel: a biological, ecological and human factors network, which were then combined to create a complete representation of free-ranging cheetah population viability. Such OOBNs have widespread relevance to the effective and targeted conservation management of vulnerable and endangered species.

Researching information experience : object and domain

Information experience has emerged as a new and dynamic field of information research in recent years. This chapter will discuss and explore information experience in two distinct ways: (a) as a research object, and; (b) as a research domain. Two recent studies will provide the context for this exploration. The first study investigated the information experiences of people using social media (e.g., Facebook, Twitter, YouTube) during natural disasters. Data was gathered by in-depth semi-structured interviews with 25 participants, from two areas affected by natural disasters (i.e., Brisbane and Townsville). The second study investigated the qualitatively different ways in which people experienced information literacy during a natural disaster. Using phenomenography, data was collected via semi-structured interviews with 7 participants. These studies represent two related yet different investigations. Taken together the studies provide a means to critically debate and reflect upon our evolving understandings of information experience, both as a research object and as a research domain. This chapter presents our preliminary reflections and concludes that further research is needed to develop and strengthen our conceptualisation of this emerging area.

Image segmentation and pose estimation of humans in video

This thesis introduces improved techniques towards automatically estimating the pose of humans from video. It examines a complete workflow to estimating pose, from the segmentation of the raw video stream to extract silhouettes, to using the silhouettes in order to determine the relative orientation of parts of the human body. The proposed segmentation algorithms have improved performance and reduced complexity, while the pose estimation shows superior accuracy during difficult cases of self occlusion.

Transit passenger segmentation using travel regularity mined from Smart Card transactions data

Transit passenger market segmentation enables transit operators to target different classes of transit users to provide customized information and services. The Smart Card (SC) data, from Automated Fare Collection system, facilitates the understanding of multiday travel regularity of transit passengers, and can be used to segment them into identifiable classes of similar behaviors and needs. However, the use of SC data for market segmentation has attracted very limited attention in the literature. This paper proposes a novel methodology for mining spatial and temporal travel regularity from each individual passenger’s historical SC transactions and segments them into four segments of transit users. After reconstructing the travel itineraries from historical SC transactions, the paper adopts the Density-Based Spatial Clustering of Application with Noise (DBSCAN) algorithm to mine travel regularity of each SC user. The travel regularity is then used to segment SC users by an a priori market segmentation approach. The methodology proposed in this paper assists transit operators to understand their passengers and provide them oriented information and services.

Advantages of exploiting projection structure for segmenting dense 3D point clouds

Timely and comprehensive scene segmentation is often a critical step for many high level mobile robotic tasks. This paper examines a projected area based neighbourhood lookup approach with the motivation towards faster unsupervised segmentation of dense 3D point clouds. The proposed algorithm exploits the projection geometry of a depth camera to find nearest neighbours which is time independent of the input data size. Points near depth discontinuations are also detected to reinforce object boundaries in the clustering process. The search method presented is evaluated using both indoor and outdoor dense depth images and demonstrates significant improvements in speed and precision compared to the commonly used Fast library for approximate nearest neighbour (FLANN) [Muja and Lowe, 2009].

Local inter-session variability modelling for object classification

Object classification is plagued by the issue of session variation. Session variation describes any variation that makes one instance of an object look different to another, for instance due to pose or illumination variation. Recent work in the challenging task of face verification has shown that session variability modelling provides a mechanism to overcome some of these limitations. However, for computer vision purposes, it has only been applied in the limited setting of face verification. In this paper we propose a local region based intersession variability (ISV) modelling approach, and apply it to challenging real-world data. We propose a region based session variability modelling approach so that local session variations can be modelled, termed Local ISV. We then demonstrate the efficacy of this technique on a challenging real-world fish image database which includes images taken underwater, providing significant real-world session variations. This Local ISV approach provides a relative performance improvement of, on average, 23% on the challenging MOBIO, Multi-PIE and SCface face databases. It also provides a relative performance improvement of 35% on our challenging fish image dataset.

CAM : a collaborative object memory system

Physical design objects such as sketches, drawings, collages, storyboards and models play an important role in supporting communication and coordination in design studios. CAM (Cooperative Artefact Memory) is a mobile-tagging based messaging system that allows designers to collaboratively store relevant information onto their design objects in the form of messages, annotations and external web links. We studied the use of CAM in a Product Design studio over three weeks, involving three different design teams. In this paper, we briefly describe CAM and show how it serves as 'object memory'.

Tracking-based moving object detection

We present a novel approach for multi-object detection in aerial videos based on tracking. The proposed method mainly involves three steps. Firstly, the spatial-temporal saliency is employed to detect moving objects. Secondly, the detected objects are tracked by mean shift in the subsequent frames. Finally, the saliency results are fused with the weight map generated by tracking to get refined detection results, and in turn the modified detection results are used to update the tracking models. The proposed algorithm is evaluated on VIVID aerial videos, and the results show that our approach can reliably detect moving objects even in challenging situations. Meanwhile, the proposed method can process videos in real time, without the effect of time delay.