Four experiments investigated matching of unfamiliar target faces taken from high-quality video against arrays of photographs. In Experiment 1, targets were present in 50% of arrays. Accuracy was poor and worsened when viewpoint and expression differed between target and array faces. In Experiment 2, targets were present in every array, but performance remained highly error prone. In Experiment 3, short video clips of the targets were shown and replayed as often as necessary, but performance levels were only slightly better than Experiment 2. Experiment 4 showed that matching was dominated by external face features. The results urge caution in the use of video images to identify people who have committed crimes. Superficial impressions of resemblance or dissimilarity between face images can be highly misleading.The human face provides the most reliable means of person identification available to the human eye (although fingerprints and iris patterns may prove more useful for automated identification; e.g., seeDaugman, 1998). Nonethe-

In this paper we propose a novel recurrent neural network architecture 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 timesteps 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.

Although temporal coding is a frequent topic of neurophysiology research, trial-to-trial variability in temporal codes is typically dismissed as noise and thought to play no role in sensory function. Here, we show that much of this supposed ''noise'' faithfully reflects stimulus-related processes carried out in coherent neural networks. Cortical neurons responded to sensory stimuli by progressing through sequences of states, identifiable only in examinations of simultaneously recorded ensembles. The specific times at which ensembles transitioned from state to state varied from trial to trial, but the state sequences were reliable and stimulusspecific. Thus, the characterization of ensemble responses in terms of state sequences captured facets of sensory processing that are missing from, and obscured in, other analyses. This work provides evidence that sensory neurons act as parts of a systems-level dynamic process, the nature of which can best be appreciated through observation of distributed ensembles.gustatory ͉ hidden Markov model T he time courses of sensory neural responses are rich with structure. Taking time into consideration increases the amount of information that can be extracted from neural codes (1-5) and changes the nature of that information (6-8). Such temporal complexity is the natural result of interactions among neural populations (9-11), a concept recently illustrated in studies of olfactory antennal lobe responses in insects (12)(13)(14).The behavior of mammalian sensory systems has proven more difficult to characterize, due in part to the relative complexity of these networks and of the behaviors and neural activity that they subtend. Feedback and convergence found in mammalian brains are extensive and diffuse (15), a fact that contributes to high trial-to-trial variability of mammalian cortical sensory responses (16). This variability is usually dismissed as noise, a decision formalized by the use of across-trial averages such as peristimulus time histograms (PSTHs) (8) and compilations of sequentially recorded neurons (13) to characterize temporal codes.If the variability in neural responses is not noise, however [if, for instance, it reflects network processes evolving at different speeds from trial to trial (17, 18)], then trial-averaging techniques will obscure features of the underlying neural processes. Recent evidence indirectly suggests that this possibility may be the case: repeating multineuronal temporal patterns that are not reflected in PSTHs follow application of sensory stimuli (19, 20) and precede initiation of motor behaviors (21-23), although the search algorithms used to identify such patterns are controversial (24, 25); furthermore, the speed of perceptual identification itself varies from trial to trial (26, 27) in a manner linked to the dynamics of network activity (27)(28)(29)(30).Here, we provide direct evidence that trial-to-trial variability is a reliable, information-rich part of ensemble sensory processing in awake rats, by using hidden Markov models [HMM (31)...

Pictures of facial expressions from the Ekman and Friesen set (Ekman, P., Friesen, W. V., (1976). Pictures of facial affect. Palo Alto, California: Consulting Psychologists Press) were submitted to a principal component analysis (PCA) of their pixel intensities. The output of the PCA was submitted to a series of linear discriminant analyses which revealed three principal findings: (1) a PCA-based system can support facial expression recognition, (2) continuous two-dimensional models of emotion (e.g. Russell, J. A. (1980). A circumplex model of affect. Journal of Personality and Social Psychology, 39, 1161-1178) are reflected in the statistical structure of the Ekman and Friesen facial expressions, and (3) components for coding facial expression information are largely different to components for facial identity information. The implications for models of face processing are discussed.

No abstract