1,721,097 research outputs found
Perp Walk Could Have Whole New Meaning
No full textScientists say an individual's walk is kind of a like a fingerprint -- no two are alike
Walk Of Shame: Criminals Exposed
No full textScientists have found a new way to identify criminals, and it is all about the way they walk. Sky's Sara Merchant met the pioneering researchers to find out how it works
Can You Identify a Criminal By His Walk?
No full textReport on Good Morning America about the gait research carried out her
A NEW APPROACH TO AUDIOVISUAL DIGITAL ARCHIVING
No full textThis paper presents work in the UK AVATAR-m project on how to specify and govern federated archive services that involve both local and remote storage
Histogram of confidences for person detection
Full text linkThis paper focuses on the problem of person detection in harsh industrial environments. Different image regions often have different requirements for the person to be detected. Additionally, as the environment can change on a frame to frame basis even previously detected people can fail to be found. In our work we adapt a previously trained classifier to improve its performance in the industrial environment. The classifier output is initially used an image descriptor. Structure from the descriptor history is learned using semi-supervised learning to boost overall performance. In comparison with two state of the art person detectors we see gains of 10%. Our approach is generally applicable to pretrained classifiers which can then be specialised for a specific scen
Reliable audiovisual archiving using unreliable storage technology and services
No full textThe drive for online access to archive content within ‘tapeless’ workflows means that mass-storage technology is an inevitable part of modern archive solutions, either in-house or provided as services by third-parties. But are these solutions safe? Can they assure the data integrity needed for long-term preservation of Petabyte volumes of data? The answer is no. Field studies reveal data corruption can take place silently without detection or correction, including in 'enterprise class' systems explicitly designed to prevent data loss. The reality is that data loss is inevitable to some degree or another from hardware failures, software bugs, and human errors. This paper presents ongoing work in the UK AVATAR-m project and in the recently started EC PrestoPrime project on a framework for storing large audiovisual files on heterogeneous and distributed storage infrastructures that allows various strategies for content replication, integrity monitoring and repair to be developed and tested
Automatic Workflow Monitoring in Industrial Environments
No full textRobust automatic workflow monitoring using visual sensors in industrial environments is still an unsolved problem. This is mainly due to the difficulties of recording data in work settings and the environmental conditions (large occlusions, similar background/foreground) which do not allow object detection/tracking algorithms to perform robustly. Hence approaches analysing trajectories are limited in such environments. However, workflow monitoring is especially needed due to quality and safety requirements. In this paper we propose a robust approach for workflow classification in industrial environments. The proposed approach consists of a robust scene descriptor and an efficient time series analysis method. Experimental results on a challenging car manufacturing dataset showed that the proposed scene descriptor is able to detect both human and machinery related motion robustly and the used time series analysis method can classify tasks in a given workflow automatically
- …
