PETS 2015: datasets and challenge

This paper presents the two datasets (ARENA and P5) and the challenge that form a part of the PETS 2015 workshop. The datasets consist of scenarios recorded by us- ing multiple visual and thermal sensors. The scenarios in ARENA dataset involve different staged activities around a parked vehicle in a parking lot in UK and those in P5 dataset involve different staged activities around the perimeter of a nuclear power plant in Sweden. The scenarios of each dataset are grouped into ‘Normal’, ‘Warning’ and ‘Alarm’ categories. The Challenge specifically includes tasks that account for different steps in a video understanding system: Low-Level Video Analysis (object detection and tracking), Mid-Level Video Analysis (‘atomic’ event detection) and High-Level Video Analysis (‘complex’ event detection). The evaluation methodology used for the Challenge includes well-established measures.

Tracking performance evaluation on PETS 2015 Challenge datasets

This paper presents a quantitative evaluation of a tracking system on PETS 2015 Challenge datasets using well-established performance measures. Using the existing tools, the tracking system implements an end-to-end pipeline that include object detection, tracking and post- processing stages. The evaluation results are presented on the provided sequences of both ARENA and P5 datasets of PETS 2015 Challenge. The results show an encouraging performance of the tracker in terms of accuracy but a greater tendency of being prone to cardinality error and ID changes on both datasets. Moreover, the analysis show a better performance of the tracker on visible imagery than on thermal imagery.

Vehicle subtype, make and model classification from side profile video

This paper addresses the challenging domain of vehicle classification from pole-mounted roadway cameras, specifically from side-profile views. A new public vehicle dataset is made available consisting of over 10000 side profile images (86 make/model and 9 sub-type classes). 5 state-of-the-art classifiers are applied to the dataset, with the best achieving high classification rates of 98.7% for sub-type and 99.7- 99.9% for make and model recognition, confirming the assertion made that single vehicle side profile images can be used for robust classification.

Meeting detection in video through semantic analysis

In this paper we present a novel approach to detect people meeting. The proposed approach works by translating people behaviour from trajectory information into semantic terms. Having available a semantic model of the meeting behaviour, the event detection is performed in the semantic domain. The model is learnt employing a soft-computing clustering algorithm that combines trajectory information and motion semantic terms. A stable representation can be obtained from a series of examples. Results obtained on a series of videos with different types of meeting situations show that the proposed approach can learn a generic model that can effectively be applied on the behaviour recognition of meeting situations.

PETS 2014: dataset and challenge

This paper describes the dataset and vision challenges that form part of the PETS 2014 workshop. The datasets are multisensor sequences containing different activities around a parked vehicle in a parking lot. The dataset scenarios were filmed from multiple cameras mounted on the vehicle itself and involve multiple actors. In PETS2014 workshop, 22 acted scenarios are provided of abnormal behaviour around the parked vehicle. The aim in PETS 2014 is to provide a standard benchmark that indicates how detection, tracking, abnormality and behaviour analysis systems perform against a common database. The dataset specifically addresses several vision challenges corresponding to different steps in a video understanding system: Low-Level Video Analysis (object detection and tracking), Mid-Level Video Analysis (‘simple’ event detection: the behaviour recognition of a single actor) and High-Level Video Analysis (‘complex’ event detection: the behaviour and interaction recognition of several actors).

A robust sclera segmentation algorithm

Sclera segmentation is shown to be of significant importance for eye and iris biometrics. However, sclera segmentation has not been extensively researched as a separate topic, but mainly summarized as a component of a broader task. This paper proposes a novel sclera segmentation algorithm for colour images which operates at pixel-level. Exploring various colour spaces, the proposed approach is robust to image noise and different gaze directions. The algorithm’s robustness is enhanced by a two-stage classifier. At the first stage, a set of simple classifiers is employed, while at the second stage, a neural network classifier operates on the probabilities’ space generated by the classifiers at stage 1. The proposed method was ranked the 1st in Sclera Segmentation Benchmarking Competition 2015, part of BTAS 2015, with a precision of 95.05% corresponding to a recall of 94.56%.

Segmentation-level fusion for iris recognition

This paper investigates the potential of fusion at normalisation/segmentation level prior to feature extraction. While there are several biometric fusion methods at data/feature level, score level and rank/decision level combining raw biometric signals, scores, or ranks/decisions, this type of fusion is still in its infancy. However, the increasing demand to allow for more relaxed and less invasive recording conditions, especially for on-the-move iris recognition, suggests to further investigate fusion at this very low level. This paper focuses on the approach of multi-segmentation fusion for iris biometric systems investigating the benefit of combining the segmentation result of multiple normalisation algorithms, using four methods from two different public iris toolkits (USIT, OSIRIS) on the public CASIA and IITD iris datasets. Evaluations based on recognition accuracy and ground truth segmentation data indicate high sensitivity with regards to the type of errors made by segmentation algorithms.

An enhanced content selection mechanism for personalization of video news programmes

In this paper, we propose a content selection framework that improves the users` experience when they are enriching or authoring pieces of news. This framework combines a variety of techniques to retrieve semantically related videos, based on a set of criteria which are specified automatically depending on the media`s constraints. The combination of different content selection mechanisms can improve the quality of the retrieved scenes, because each technique`s limitations are minimized by other techniques` strengths. We present an evaluation based on a number of experiments, which show that the retrieved results are better when all criteria are used at time.

Selecting salient objects in real scenes: An oscillatory correlation model

Attention is a critical mechanism for visual scene analysis. By means of attention, it is possible to break down the analysis of a complex scene to the analysis of its parts through a selection process. Empirical studies demonstrate that attentional selection is conducted on visual objects as a whole. We present a neurocomputational model of object-based selection in the framework of oscillatory correlation. By segmenting an input scene and integrating the segments with their conspicuity obtained from a saliency map, the model selects salient objects rather than salient locations. The proposed system is composed of three modules: a saliency map providing saliency values of image locations, image segmentation for breaking the input scene into a set of objects, and object selection which allows one of the objects of the scene to be selected at a time. This object selection system has been applied to real gray-level and color images and the simulation results show the effectiveness of the system. (C) 2010 Elsevier Ltd. All rights reserved.

On modeling and evaluating multicomputer transcoding architectures for live-video streams

The pervasive and ubiquitous computing has motivated researches on multimedia adaptation which aims at matching the video quality to the user needs and device restrictions. This technique has a high computational cost which needs to be studied and estimated when designing architectures and applications. This paper presents an analytical model to quantify these video transcoding costs in a hardware independent way. The model was used to analyze the impact of transcoding delays in end-to-end live-video transmissions over LANs, MANs and WANs. Experiments confirm that the proposed model helps to define the best transcoding architecture for different scenarios.

Watch-and-Comment as a Paradigm toward Ubiquitous Interactive Video Editing

The literature reports research efforts allowing the editing of interactive TV multimedia documents by end-users. In this article we propose complementary contributions relative to end-user generated interactive video, video tagging, and collaboration. In earlier work we proposed the watch-and-comment (WaC) paradigm as the seamless capture of an individual`s comments so that corresponding annotated interactive videos be automatically generated. As a proof of concept, we implemented a prototype application, the WACTOOL, that supports the capture of digital ink and voice comments over individual frames and segments of the video, producing a declarative document that specifies both: different media stream structure and synchronization. In this article, we extend the WaC paradigm in two ways. First, user-video interactions are associated with edit commands and digital ink operations. Second, focusing on collaboration and distribution issues, we employ annotations as simple containers for context information by using them as tags in order to organize, store and distribute information in a P2P-based multimedia capture platform. We highlight the design principles of the watch-and-comment paradigm, and demonstrate related results including the current version of the WACTOOL and its architecture. We also illustrate how an interactive video produced by the WACTOOL can be rendered in an interactive video environment, the Ginga-NCL player, and include results from a preliminary evaluation.

Chaotic synchronization in general network topology for scene segmentation

Chaotic synchronization has been discovered to be an important property of neural activities, which in turn has encouraged many researchers to develop chaotic neural networks for scene and data analysis. In this paper, we study the synchronization role of coupled chaotic oscillators in networks of general topology. Specifically, a rigorous proof is presented to show that a large number of oscillators with arbitrary geometrical connections can be synchronized by providing a sufficiently strong coupling strength. Moreover, the results presented in this paper not only are valid to a wide class of chaotic oscillators, but also cover the parameter mismatch case. Finally, we show how the obtained result can be applied to construct an oscillatory network for scene segmentation.

Chaotic synchronization in 2D lattice for scene segmentation

Synchronization and chaos play important roles in neural activities and have been applied in oscillatory correlation modeling for scene and data analysis. Although it is an extensively studied topic, there are still few results regarding synchrony in locally coupled systems. In this paper we give a rigorous proof to show that large numbers of coupled chaotic oscillators with parameter mismatch in a 2D lattice can be synchronized by providing a sufficiently large coupling strength. We demonstrate how the obtained result can be applied to construct an oscillatory network for scene segmentation. (C) 2007 Elsevier B.V. All rights reserved.

An entropy-based approach to automatic image segmentation of satellite images

An entropy-based image segmentation approach is introduced and applied to color images obtained from Google Earth. Segmentation refers to the process of partitioning a digital image in order to locate different objects and regions of interest. The application to satellite images paves the way to automated monitoring of ecological catastrophes, urban growth, agricultural activity, maritime pollution, climate changing and general surveillance. Regions representing aquatic, rural and urban areas are identified and the accuracy of the proposed segmentation methodology is evaluated. The comparison with gray level images revealed that the color information is fundamental to obtain an accurate segmentation. (C) 2010 Elsevier B.V. All rights reserved.

Cross-situational learning of object-word mapping using Neural Modeling Fields

The issue of how children learn the meaning of words is fundamental to developmental psychology. The recent attempts to develop or evolve efficient communication protocols among interacting robots or Virtual agents have brought that issue to a central place in more applied research fields, such as computational linguistics and neural networks, as well. An attractive approach to learning an object-word mapping is the so-called cross-situational learning. This learning scenario is based on the intuitive notion that a learner can determine the meaning of a word by finding something in common across all observed uses of that word. Here we show how the deterministic Neural Modeling Fields (NMF) categorization mechanism can be used by the learner as an efficient algorithm to infer the correct object-word mapping. To achieve that we first reduce the original on-line learning problem to a batch learning problem where the inputs to the NMF mechanism are all possible object-word associations that Could be inferred from the cross-situational learning scenario. Since many of those associations are incorrect, they are considered as clutter or noise and discarded automatically by a clutter detector model included in our NMF implementation. With these two key ingredients - batch learning and clutter detection - the NMF mechanism was capable to infer perfectly the correct object-word mapping. (C) 2009 Elsevier Ltd. All rights reserved.