Data-Augmentation for Reducing Dataset Bias in Person Re-identification

In this paper we explore ways to address the issue of dataset bias in person re-identification by using data augmentation to increase the variability of the available datasets, and we introduce a novel data augmentation method for re-identification based on changing the image background. We show that use of data augmentation can improve the cross-dataset generalisation of convolutional network based re-identification systems, and that changing the image background yields further improvements.

Recurrent Convolutional Network for Video-based Person Re-Identification

In this paper we propose a novel recurrent neural networkarchitecture for video-based person re-identification.Given the video sequence of a person, features are extracted from each frame using a convolutional neural network that incorporates a recurrent final layer, which allows information to flow between time-steps. The features from all time steps are then combined using temporal pooling to give an overall appearance feature for the complete sequence. The convolutional network, recurrent layer, and temporal pooling layer, are jointly trained to act as a feature extractor for video-based re-identification using a Siamese network architecture.Our approach makes use of colour and optical flow information in order to capture appearance and motion information which is useful for video re-identification. Experiments are conduced on the iLIDS-VID and PRID-2011 datasets to show that this approach outperforms existing methods of video-based re-identification.

https://github.com/niallmcl/Recurrent-Convolutional-Video-ReID
Project Source Code

Person Re-Identification using Deep Convnets with Multi-task Learning

Person re-identification involves recognizing a person across non-overlapping camera views, with different pose, illumination, and camera characteristics. We propose to tackle this problem by training a deep convolutional network to represent a person’s appearance as a low-dimensional feature vector that is invariant to common appearance variations encountered in the re-identification problem. Specifically, a Siamese-network architecture is used to train a feature extraction network using pairs of similar and dissimilar images. We show that use of a novel multi-task learning objective is crucial for regularizing the network parameters in order to prevent over-fitting due to the small size the training dataset. We complement the verification task, which is at the heart of re-identification, by training the network to jointly perform verification, identification, and to recognise attributes related to the clothing and pose of the person in each image. Additionally, we show that our proposed approach performs well even in the challenging cross-dataset scenario, which may better reflect real-world expected performance. 

Intrinsic variability in the detection of micrometastases in lymph nodes for re-staging of colorectal cancer: effect of individual markers and tissue samples

In this study, we investigated whether (a) carcinoembryonic antigen (CEA), cytokeratin-20 (CK-20) and guanylyl cyclase C (GCC) are clinically useful markers for the molecular detection of submicroscopic metastases in colorectal cancer (CRC) and (b) whether overexpression of CEA, CK-20 and GCC can be reliably detected in formalin-fixed, paraffin-embedded tissues as well as frozen lymph nodes. We studied 175 frozen lymph nodes and 158 formalin-fixed, paraffin-embedded lymph nodes from 28 cases of CRC. CEA or CK-20 or GCC-specific polymerase chain reaction (PCR) was carried out on mRNA transcripts extracted from the nodal tissues. Ten out of I I Dukes' B CRC cases had detectable CEA and CK-20 while 6 out of 11 Dukes' B CRC cases had detectable GCC. In general, the difference of re-staged cases when comparing frozen and paraffin-embedded samples was marked; the only statistically significant correlation between frozen and paraffin tissue was for the CEA marker. Our results indicated a high incidence (>50%) of detecting micrometastases in histologically-negative lymph nodes at the molecular level. (C) 2003 Elsevier Science Ltd. All rights reserved.

Device-Free Person Detection and Ranging in UWB Networks

We present a novel device-free stationary person detection and ranging method, that is applicable to ultra-wide bandwidth (UWB) networks. The method utilizes a fixed UWB infrastructure and does not require a training database of template waveforms. Instead, the method capitalizes on the fact that a human presence induces small low-frequency variations that stand out against the background signal, which is mainly affected by wideband noise. We analyze the detection probability, and validate our findings with numerical simulations and experiments with off-the-shelf UWB transceivers in an indoor environment. © 2007-2012 IEEE.

An experimental study of UWB device-free person detection and ranging

Passive person detection and localization is an emerging area in UWB localization systems, whereby people are not required to carry any UWB ranging device. Based on experimental data, we propose a novel method to detect static persons in the absence of template waveforms, and to compute distances to these persons. Our method makes very little assumptions on the environment and can achieve ranging performances on the order of 50 cm, using off-the-shelf UWB devices. © 2013 IEEE.

Detection of cathepsin S cysteine protease in human brain tumour microdialysates in vivo.

Microdialysis enables the chemistry of extracellular ?uid in body tissues to be measured. Extracellular proteases such as the cysteine protease, cathepsin S (CatS), are thought to facilitate astrocytoma invasion. Microdialysates obtained from human brain tumoursin vivo were subjected to cathepsin S activity and ELISA assays. Cathepsin S ELISA expression was detected in ?ve out of 10 tumour microdialysates, while activity was detected in ?ve out of 11 tumour microdialysates. Cathepsin S expression was also detected in microdialysate from the normal brain control although no activity was found in the same sample. While some re?nements to the technique are necessary, the authors demonstrate the feasibility and safety of microdialysis in human astrocytomasin vivo. Characterisation of the extracellular environment of brain tumoursin vivo using microdialysis may be a useful tool to identify the protease pro?le of brain tumours.

Real-time error detection but not error correction drives automatic visuomotor adaptation

We investigated the role of visual feedback of task performance in visuomotor adaptation. Participants produced novel two degrees of freedom movements (elbow flexion-extension, forearm pronation-supination) to move a cursor towards visual targets. Following trials with no rotation, participants were exposed to a 60A degrees visuomotor rotation, before returning to the non-rotated condition. A colour cue on each trial permitted identification of the rotated/non-rotated contexts. Participants could not see their arm but received continuous and concurrent visual feedback (CF) of a cursor representing limb position or post-trial visual feedback (PF) representing the movement trajectory. Separate groups of participants who received CF were instructed that online modifications of their movements either were, or were not, permissible as a means of improving performance. Feedforward-mediated performance improvements occurred for both CF and PF groups in the rotated environment. Furthermore, for CF participants this adaptation occurred regardless of whether feedback modifications of motor commands were permissible. Upon re-exposure to the non-rotated environment participants in the CF, but not PF, groups exhibited post-training aftereffects, manifested as greater angular deviations from a straight initial trajectory, with respect to the pre-rotation trials. Accordingly, the nature of the performance improvements that occurred was dependent upon the timing of the visual feedback of task performance. Continuous visual feedback of task performance during task execution appears critical in realising automatic visuomotor adaptation through a recalibration of the visuomotor mapping that transforms visual inputs into appropriate motor commands.

Detection of aberrant behaviour in home environments from video sequence

We introduce an application for the detection of aberrant behaviour within home based environments, with a focus on repetitive actions, which may be present in instance of persons suffering from dementia. Video based analysis has been used to detect the motion of a person within a given scene in addition to tracking them over the time. Detection of repetitive actions has been based on the analysis of a person's trajectory using the principles of signal correlation. Along with the ability to detect repetitive motion the developed approach also has the ability to measure the amount of activity/inactivity within the scene during a given period of time. Our results showed that the developed approach had the ability to detect all patterns in the data set examined with an average accuracy of 96.67%. This work has therefore validated the proposed concept of video based analysis for the detection of repetitive activities.

Terahertz detection of illegal objects

This paper describes the main parameters - contrast, spatial resolution, and thermal sensitivity - which define the performance of any stand-off imaging system. The origin of the signature for both metal and dielectric objects hidden under clothing in the frequency range from 100 GHz to 500 GHz is discussed. At 100 GHz the signature is dominated by reflection whilst at 500 GHz it is dominated by emission. A 94-GHz-passive millimetre-wave imaging system has been designed and fabricated to image objects under clothing. This imager is based on a Schmidt camera folded using polarisation techniques.

Detection of undiagnosed coeliac disease with atypical features using antireticulin and antigliadin antibodies

Eighteen patients with a variety of non-gastrointestinal symptoms were incidentally found to have circulating antireticulin antibody and on subsequent testing were also positive for antigliadin antibody. They prospectively underwent jejunal biopsy to determine whether or not they had coeliac disease. Their age range was 21-79 years (mean 42 years). Enteropathy was present in 13 (72 per cent) and was always associated with circulating IgA antigliadin antibody. Enteropathy was not present in the five cases who had only IgG antibody. Clinical improvement occurred in eight of 11 patients who complied with a gluten-free diet and was paralleled by an improvement in the mucosal histology in seven of eight who were re-biopsied. The most remarkable cases were two patients who presented with severe debility and no apparent haematological or biochemical abnormalities, and who subsequently made a dramatic recovery on a gluten-free diet. It is concluded that antireticulin antibody detected by routine autoantibody screening and confirmed to have IgA antigliadin antibody specificity is a useful indicator of an otherwise undiagnosed enteropathy. This serves to emphasize that the condition can sometimes be associated with atypical features and significant morbidity.

Direct Detection of YORP Spin-up on Asteroid (25143) Itokawa and Implications for its Internal Structure

Near-Earth asteroid (25143) Itokawa was visited by the Hayabusa spacecraft in 2005, resulting in a highly detailed surface shape and topography model. This model has led to several predictions for the expected radiative torques on this asteroid, suggesting that its spin rate should be decelerating. Through an observational survey spanning 2001 to 2013 we have successfully measured an acceleration in its spin rate of dω/dt = 3.54 (± 0.38) × 10^(-8) rad day^(-2), equivalent to a decrease of its rotation period of ~ 45 ms year^(-1). Using the shape model determined from the Hayabusa spacecraft, we applied a detailed thermophysical analysis, to reconcile the predicted YORP strength with that observed. We find that the center-of-mass for Itokawa must be shifted by ~20 m along the long-axis of the asteroid to reconcile observations with theory. This can be explained if Itokawa is composed of two separate bodies with very different bulk densities of 1740 ± 110 kg m^(-3) and 2730 ± 440 kg m^(-3), and was formed from the merger of two separate bodies, consistent with the collapse of a binary system or the re-accumulation of material from a catastrophic collisional disruption. We demonstrate that an observational measurement of radiative torques, when combined with a detailed shape model, can provide insight into the interior structure of an asteroid.