This research project proposes to advance state-of-the-art image recognition techniques to be able to recognize a large number ofscenes and object categories in real and unconstrained indoor andoutdoor environments i.e. traffic scenes (cars, bicycle vehicles,pedestrians, human faces, street signs etc.), urban and naturalscenes (buildings, landscapes etc.) with various rigid andarticulated objects as well as textures. Nowadaysalmost everybody carries a digital camera and taking a photo or ashort video has never been easier. Broadcasting companies receivethousands of pictures from the general public after every majorevent and the annotation of those documents is done manually. Crime investigators collect large amounts ofvisual evidence and its classification is also done manually. The UKhas the largest number of security cameras in Europe but the dataprovided by the cameras is very little explored. Furthermore,recognition and interpretation of visual information is one of themajor requirements for autonomous intelligent robots. There is therefore a dire need for a reliable recognition system capable of automatic classification and annotation of large amounts of visual documents. Any success towards achieving that goal i.e., automatic prioritizing of document browsing for experts, will be seen as a clear benefit in improvingthe efficiency of work.To fulfil the objectives of this project major progress has to bemade in the domain of features extraction, category representationand efficient search. Recent interest point based approachesdemonstrate the capability of dealing with large numbers ofcategories in the context of visual recognition. These methods showpromising directions towards successful scene and objectrecognition. Based on these results we propose to develop noveltechniques for extracting image features robust to backgroundclutter and viewpoint change, which are currently great challengesin image recognition domain. Those features will be suitable forsimultaneous representation of scenes and objects at variousappearance and structure levels as well as for segmentation ofobjects. Mid-level image segmentation methods have a potential toprovide such features and can bridge the gap between interest pointdetectors and semantic segmentation in the context of categoryrecognition. There has been little overlap between recognition andsegmentation domains although the goal is to solve both problemssimultaneously.We also propose to introduce novel hierarchical representationswhich will exploit the properties of new features and allow to dealefficiently with large number of image categories. Therepresentation will model the categories in multiple hierarchies ofvarious image attributes i.e., intensity, color and texture as wellas relations between different object parts and views. The multiplehierarchies will allow for coarse-to-fine classification based onimage cues relevant to the query. Very little work has been done inthis area and the proposed research can shed new light on imagerepresentation problems. Finally, efficient tree structures andnearest neighbor search techniques will be employed to handle largeamounts of data in multi-category learning.Developing novel, efficient and robust techniques which may providesuccessful solutions to fundamental recognition problems and advancethe state-of-the-art in feature extraction, categoryrepresentation and data exploration, make this project verychallenging and adventurous. The project is expected to achieve theobjectives within 36 months and it will involve a research student,a research assistant and the principal investigator.

The core research team comprises four academics (Dr Claire Wardle, Professor Justin Lewis, Dr Karin Wahl-Jorgensen and Dr Tammy Boyce) based at the Cardiff School of Journalism, Media and Cultural Studies, (all of whom have published in the area of citizenship, engagement and news) and a Senior Community Producer/Broadcast Journalist at the BBC (Robin Hamman), who has a strong track record publishing about online communities in academic and journalistic publications. The core team will be supported and advised by the Editor of English Regions and New Media (Liz Howell), Editor of New Media, BBC Wales (Iain Tweedale) and Head of News and Current Affairs, BBC Wales (Mark O'Callahan). The project will definitely undertake research in London, Birmingham and Cardiff, and possibly other locations if time and money allow.\n\nIn order to answer these three main objectives, the research will have four different elements\n\nStage 1: Audience Study \nWe will conduct:\na. A quantitative electronic survey of current users. Viewers who contact the BBC via email, the website or SMS will receive an automatic response asking them to fill in a short questionnaire. \nb. A quantitative survey of non-users. A representative survey of the British population will be carried out to find out who contributes to UGC, why other people do not, and how they feel about UGC in BBC news and current affairs formats.\nc. Qualitative interviews and focus groups with users and non-users about their motivations for participating, experiences of participating, and opinions about the UGC which is included in BBC output.\n\nStage 2: Content Study\nWe will conduct a content analysis of the UGC submitted to a sample of BBC formats from: Regional and National news pages; the local 'Where I Live' pages in Cardiff, Birmingham and a third site [likely Manchester or Leeds]; and high profile network radio and TV programming (including, where possible, BBC Breakfast News; Newsnight; and Radio 4's Today programme). This will provide a precise snapshot of how current BBC practices are manifested in their output.\n\nStage 3: Production Study\nWe will use unstructured interviews and participant observation to discover how UGC is perceived and used in different BBC locations, both nationally and regionally. Researchers would spend time in the newsrooms chosen for the Content Study. \n\nStage 4: Future Directions\nExperimental research to test different UGC formats to discover which are more popular with audiences and which encourages the most learning and engagement with the material. The same news story will be told using a traditional news format, a current UGC format and three different innovative UGC formats. Audience reactions will be compared.\n\nThis research project will enable the BBC to understand with greater sophistication why some people provide UGC, why others do not, and what audiences in general think of the increasing use of UGC in BBC output. The research will also provide concrete guidelines for using UGC in the newsroom, considering how to widen participation, improve cost-effectiveness, minimise risk for contributors, and maximise and streamline the flow of UGC throughout the organisation. The research project will also produce presentation materials which can be used by the BBC for in-house training. Finally, the research will create pilot UGC formats which will be designed in conjunction with focus groups, and tested on different audiences to provide the BBC with evidence about which formats encourage the most participation, and which formats are most popular with audiences in general. For the academic community, this project will provide a data set for understanding the phenomenon of UGC more clearly, both from the production and audience perspectives. \n \n

The BBC's future audience is ageing. Yet older customers are less likely to exploit advanced digital services: this project will use interview and experiment to investigate why this is so. In particular, the project will consider the reasons for 'inhibited exploration' - a reduced tendency to 'try things out'. \n\nThe project will investigate two independent causes for inhibited exploration. The first relates to older viewers' diminishing intellectual capabilities, and their tendency to change behaviour so as to reduce the load on these capabilities. Exploration loads memory (so as to keep track of what has already been tried, and to distinguish things that have been tried recently from things that have been tried on another occasion). Older viewers will find these memory tasks more difficult, and may choose to avoid them, particularly if they fear reaching 'catastrophic states' of the technology / screens from which they can't return to their tried-and-tested services.\n\nMuch interface design effort has attempted to address these issues (both in BBC and elsewhere), and so it becomes important to investigate how severe and important this issue remains. It is also important to try to discover aspects of older viewers' dispositions and situations that may ameliorate the problem.\n\nThe second cause of inhibited exploration is that older viewers may find advanced services unattractive. This may appear to be a failure of programme making or service provision; however, recent work on the psychology of decision making has shown that this issue is not simple. When people choose to try a new service, or watch a new TV program, they do so on the basis of their prediction of what they will enjoy. But how are they to do this? It's easy to predict whether or not you will enjoy something that you have consumed many times, but new products are more challenging. In fact, a body of experimental work has shown systematic discrepancies between the prediction of pleasure and its actual experience. Similarly, there are notable discrepancies between pleasure measured as it happens (imagine asking someone during a TV programme how much they are enjoying it) and pleasure measured retrospectively, from memory. The relations between these predictions and measures of enjoyment are crucial for the consumption of new products, including interactive digital services. However, most work in this area has used young participants (students), so it remains unclear what particular issues are salient in the case of older populations. \n\nThe first phase of the project will use in situ interviews to explore these issues. The interviews will also target an analysis of each cause in terms the situational and dispositional parameters. Are there some older viewers who are invulnerable to the fear of exploration / and if so why? What is the role of social support? (It is well-known that social support changes many aspects of older people's behaviour and appraisals, and it seems plausible that social support will ameliorate the fear of reaching catastrophic states, or encourage the appreciation of what is to be gained from novel services.)\n\nOnce the interviews have been analysed we will have a fuller appreciation of the nature of inhibited exploration, and we will also have a set of cases or stories that illustrate its operation, and how it relates to features of services and interfaces. These case-studies will be packaged as illustrative design scenarios to capture good and bad design practice.\n\nThe understanding gained from the interviews will allow us to prioritise issues for the second phase, for experimental studies. Experiments on exploration will evaluate interface design solutions, using mockups of the solutions already designed by BBC Future Media & Technology Division. Experiments on predicted and experienced enjoyment will use new content to explore the relations among predicted, momentary and retrospective judgments of TV content.

Many of us now carry around technologies which allow us to record sound, whether that is the sound of our child's first music concert on a digital camera or a recording of a practical joke on a mobile phone. Nowadays, there are many outlets for this user generated content. Last year alone, 13 million hours of video was uploaded to YouTube. Even professional broadcasters rely on this footage. Mainstream news bulletins regularly use amateur footage of dramatic events (e.g. Concorde crashing) and some TV programmes such as Rude Tube are entirely made up of user generated content. However, the production quality of the sound on user-generated content is often very poor: distorted, noisy, with garbled speech or indistinct music. Our interest lies in the causes of the poor recording, especially what happens between the sound source and the electronic signal emerging from the microphone. Typical problems include: speaking off microphone; distorted speech due to clipping; wind noise and microphone handling noise. We are interested in audio recorded on its own, as well as soundtracks accompanying videos. We want to improve the recording quality so that more user-generated audio can be widely used and re-used creatively. To do this we will develop an understanding of how recording errors are perceived as it is unclear how noise and distortion affects the perception of the audio quality for many sounds. We will develop algorithms for automatically evaluating audio quality from the poor recording. A method for evaluating recorded audio quality has many potential uses. When media is received by a broadcast organisation, whether submitted by an amateur or professional, a rapid quality assessment could determine whether the sound is of broadcast quality without time consuming auditioning. Searching for sounds on the Internet for creative re-use is a frustrating activity as it is difficult to find recordings, and those that are found are often of poor quality. An audio quality assessment method would make it possible to tag and search sound files for content and quality. Even better, it would be possible to use the audio quality rating at the time of recording to try and improve the quality of the captured sound. A simple warning displayed on the recording device would give an opportunity to correct mistakes (a warning light when someone is being recorded off-mic). Furthermore, a rating of audio quality could be used to produce devices which automatically correct common recording errors. The medium term aim of this research is to develop such algorithms to correct common recording errors, however, a pre-requisite is a method by which the quality of audio can be evaluated. And so that is the focus of this proposed project.

It is currently a very exciting and challenging time for video compression. The predicted growth in demand for bandwidth, especially for mobile services is driven largely by video applications and is probably greater now than it has ever been. There are four reasons for this: (i) Recently introduced formats such as 3D and multiview, coupled with increasing dynamic range, spatial resolution and framerate, all require increased bit-rate to deliver improved immersion; (ii) Video-based web traffic continues to grow and dominate the internet; (iii) User expectations coninue to drive flexibility and quality, with a move from linear to non-linear delivery; (iv) Finally the emergence of new services, in particular mobile delivery through 4G/LTE to smart phones. While advances in network and physical layer technologies will no doubt contribute to the solution, the role of video compression is also of key importance. This research project is underpinned by the assumption that, in most cases, the target of video compression is to provide good subjective quality rather than to minimise the error between the original and coded pictures. It is thus possible to conceive of a compression scheme where an analysis/synthesis framework replaces the conventional energy minimisation approach. Such a scheme could offer substantially lower bitrates through reduced residual and motion vector coding. The approach proposed will model scene content using combinations of waveform coding and texture replacement, using computer graphic models to replace target textures at the decoder. These not only offer the potential for dramatic improvements in performance, but they also provide an inherent content-related parameterisation which will be of use in classification and detection tasks as well as facilitating integration with CGI. This has the potential to create a new content-driven framework for video compression. In this context our aim is to shift the video coding paradigm from rate-distortion optimisation to rate-quality modelling, where region-based parameters are combined with perceptual quality metrics to inform and drive the coding and synthesis processes. However it is clear that a huge amount of research needs to be done in order to fully exploit the method's potential and to yield stable and efficient solutions. For example, mean square error is no longer a valid objective function or measure of quality, and new embedded perceptually driven quality metrics are essential. The choice of texture analysis and synthesis models are also important, as is the exploitation of long-term picture dependencies.