The Experts below are selected from a list of 219 Experts worldwide ranked by ideXlab platform

Marta Chrustowicz - One of the best experts on this subject based on the ideXlab platform.

  • Social Perception and Interaction Database-A Novel Tool to Study Social Cognitive Processes With Point-Light Displays.
    Frontiers in psychiatry, 2020
    Co-Authors: Łukasz Okruszek, Marta Chrustowicz
    Abstract:

    Introduction: The ability to detect and interpret social interactions (SI) is one of the crucial skills enabling people to operate in the social world. Multiple lines of evidence converge to indicate the preferential processing of SI when compared to the individual actions of multiple agents, even if the actions were visually degraded to minimalistic point-light displays (PLDs). Here, we present a novel PLD dataset (Social Perception and Interaction Database; SoPID) that may be used for studying multiple levels of social information processing. Methods: During a motion-Capture Session, two pairs of actors were asked to perform a wide range of 3-second actions, including: (1) neutral, gesture-based communicative interactions (COM); (2) emotional exchanges (Happy/Angry); (3) synchronous interactive physical activity of actors (SYNC); and (4) independent actions of agents, either object-related (ORA) or non-object related (NORA). An interface that allows single/dyadic PLD stimuli to be presented from either the second person (action aimed towards the viewer) or third person (observation of actions presented toward other agents) perspective was implemented on the basis on the recorded actions. Two validation studies (each with 20 healthy individuals) were then performed to establish the recognizability of the SoPID vignettes. Results: The first study showed a ceiling level accuracy for discrimination of communicative vs. individual actions (93% +/- 5%) and high accuracy for interpreting specific types of actions (85 +/- 4%) from the SoPID. In the second study, a robust effect of scrambling on the recognizability of SoPID stimuli was observed in an independent sample of healthy individuals. Discussion: These results suggest that the SoPID may be effectively used to examine processes associated with communicative interactions and intentions processing. The database can be accessed via the Open Science Framework (https://osf.io/dcht8/).

  • Social Perception and Interaction Database – a novel tool to study social cognitive processes with point-light displays
    2019
    Co-Authors: Lukasz Okruszek, Marta Chrustowicz
    Abstract:

    Introduction: The ability to detect and interpret third-party encounters (TPE) is one of the crucial skills enabling people to operate in the social world. Multiple lines of evidence converge towards the preferential processing of TPE when compared to the individual actions of multiple agents, even if the actions of agents were visually degraded to minimalistic point-light displays (PLDs). Here, we present a novel PLD dataset (Social Perception and Interaction Database; SoPID) that may be used for studying multiple levels of social information processing. Methods: During a motion-Capture Session, two pairs of actors were asked to perform a wide range of dyadic 3-second actions, including: (1) neutral, gesture-based communicative interactions (COM); (2) emotional exchanges (Happy/Angry); (3) synchronous physical activity of actors (SYNC); and (4) independent actions of agents, either object-related (ORA) or non-object related (NORA). The stimuli were then transformed into PLDs. Two validation studies (each with 20 healthy individuals) were then performed to establish the recognizability of the SoPID vignettes. Results: The first study showed a ceiling level accuracy for discrimination of communicative vs. individual actions (93% +/- 5%) and high accuracy for interpreting specific types of actions (85 +/- 4%) from the SoPID. In the second study, a robust effect of scrambling on the recognizability of SoPID vignettes was observed in an independent sample of healthy individuals. Discussion: These results suggest that the SoPID may be effectively used to examine processes associated with communicative interactions and intentions processing. The database can be accessed via Open Science Framework (https://osf.io/dcht8/).

Łukasz Okruszek - One of the best experts on this subject based on the ideXlab platform.

  • Social Perception and Interaction Database-A Novel Tool to Study Social Cognitive Processes With Point-Light Displays.
    Frontiers in psychiatry, 2020
    Co-Authors: Łukasz Okruszek, Marta Chrustowicz
    Abstract:

    Introduction: The ability to detect and interpret social interactions (SI) is one of the crucial skills enabling people to operate in the social world. Multiple lines of evidence converge to indicate the preferential processing of SI when compared to the individual actions of multiple agents, even if the actions were visually degraded to minimalistic point-light displays (PLDs). Here, we present a novel PLD dataset (Social Perception and Interaction Database; SoPID) that may be used for studying multiple levels of social information processing. Methods: During a motion-Capture Session, two pairs of actors were asked to perform a wide range of 3-second actions, including: (1) neutral, gesture-based communicative interactions (COM); (2) emotional exchanges (Happy/Angry); (3) synchronous interactive physical activity of actors (SYNC); and (4) independent actions of agents, either object-related (ORA) or non-object related (NORA). An interface that allows single/dyadic PLD stimuli to be presented from either the second person (action aimed towards the viewer) or third person (observation of actions presented toward other agents) perspective was implemented on the basis on the recorded actions. Two validation studies (each with 20 healthy individuals) were then performed to establish the recognizability of the SoPID vignettes. Results: The first study showed a ceiling level accuracy for discrimination of communicative vs. individual actions (93% +/- 5%) and high accuracy for interpreting specific types of actions (85 +/- 4%) from the SoPID. In the second study, a robust effect of scrambling on the recognizability of SoPID stimuli was observed in an independent sample of healthy individuals. Discussion: These results suggest that the SoPID may be effectively used to examine processes associated with communicative interactions and intentions processing. The database can be accessed via the Open Science Framework (https://osf.io/dcht8/).

Prakhar Gupta - One of the best experts on this subject based on the ideXlab platform.

  • ECIR (1) - Using Image Captions and Multitask Learning for Recommending Query Reformulations.
    Lecture Notes in Computer Science, 2020
    Co-Authors: Gaurav Verma, Vishwa Vinay, Sahil Bansal, Shashank Oberoi, Makkunda Sharma, Prakhar Gupta
    Abstract:

    Interactive search Sessions often contain multiple queries, where the user submits a reformulated version of the previous query in response to the original results. We aim to enhance the query recommendation experience for a commercial image search engine. Our proposed methodology incorporates current state-of-the-art practices from relevant literature – the use of generation-based sequence-to-sequence models that Capture Session context, and a multitask architecture that simultaneously optimizes the ranking of results. We extend this setup by driving the learning of such a model with captions of clicked images as the target, instead of using the subsequent query within the Session. Since these captions tend to be linguistically richer, the reformulation mechanism can be seen as assistance to construct more descriptive queries. In addition, via the use of a pairwise loss for the secondary ranking task, we show that the generated reformulations are more diverse.

  • Using Image Captions and Multitask Learning for Recommending Query Reformulations
    arXiv: Information Retrieval, 2020
    Co-Authors: Gaurav Verma, Vishwa Vinay, Sahil Bansal, Shashank Oberoi, Makkunda Sharma, Prakhar Gupta
    Abstract:

    Interactive search Sessions often contain multiple queries, where the user submits a reformulated version of the previous query in response to the original results. We aim to enhance the query recommendation experience for a commercial image search engine. Our proposed methodology incorporates current state-of-the-art practices from relevant literature -- the use of generation-based sequence-to-sequence models that Capture Session context, and a multitask architecture that simultaneously optimizes the ranking of results. We extend this setup by driving the learning of such a model with captions of clicked images as the target, instead of using the subsequent query within the Session. Since these captions tend to be linguistically richer, the reformulation mechanism can be seen as assistance to construct more descriptive queries. In addition, via the use of a pairwise loss for the secondary ranking task, we show that the generated reformulations are more diverse.

Andrew W Byrne - One of the best experts on this subject based on the ideXlab platform.

  • population estimation and trappability of the european badger meles meles implications for tuberculosis management
    PLOS ONE, 2012
    Co-Authors: Andrew W Byrne, James Okeeffe, Stuart Green, Paddy D Sleeman, L A Corner, Eamonn Gormley, Denise Murphy, Wayne S Martin, John Davenport
    Abstract:

    Estimates of population size and trappability inform vaccine efficacy modelling and are required for adaptive management during prolonged wildlife vaccination campaigns. We present an analysis of mark-reCapture data from a badger vaccine (Bacille Calmette-Guerin) study in Ireland. This study is the largest scale (755 km 2 ) mark-reCapture study ever undertaken with this species. The study area was divided into three approximately equal-sized zones, each with similar survey and Capture effort. A mean badger population size of 671 (SD: 76) was estimated using a closed-subpopulation model (CSpM) based on data from capturing Sessions of the entire area and was consistent with a separate multiplicative model. Minimum number alive estimates calculated from the same data were on average 49-51% smaller than the CSpM estimates, but these are considered severely negatively biased when trappability is low. Population densities derived from the CSpM estimates were 0.82-1.06 badgers km 22 , and broadly consistent with previous reports for an adjacent area. Mean trappability was estimated to be 34-35% per Session across the population. By the fifth Capture Session, 79% of the adult badgers caught had been marked previously. Multivariable modelling suggested significant differences in badger trappability depending on zone, season and age-class. There were more putatively trap-wary badgers identified in the population than trap-happy badgers, but wariness was not related to individual's sex, zone or season of Capture. Live-trapping efficacy can vary significantly amongst sites, seasons, age, or personality, hence monitoring of trappability is recommended as part of an adaptive management regime during large-scale wildlife vaccination programs to counter biases and to improve efficiencies.

Richard Guest - One of the best experts on this subject based on the ideXlab platform.

  • Age dependency in handwritten dynamic signature verification systems
    Pattern Recognition Letters, 2006
    Co-Authors: Richard Guest
    Abstract:

    An assessment of age dependency of biometric dynamic signature verification systems is presented. A number of commonly used features are extracted from multiple signature sample donated by 274 signers. These features are examined for repeatability both within a single signature Capture Session and between multiple Sessions with particular consideration for any performance variation between age groups. Alongside this analysis, an age evaluation of test subjects' ability to enrol/validate on standard systems is presented. Performance is stable across all evaluated age groups proving the ability signature system to be deployed for use within a general population. Some performance features however vary significantly with age in terms of repeatability and mean feature value; characteristics which can be exploited and must be accommodated in the design of systems for use amongst a wide or specific population.

  • IWFHR - The repeatability of signatures
    Ninth International Workshop on Frontiers in Handwriting Recognition, 1
    Co-Authors: Richard Guest
    Abstract:

    Signatures are the most widely used form of legally binding identification and authentication. The repeatability of a person's signature underpins its recognition and hence usefulness in everyday authentication situations. This study aims to assess the stability of a set of common features used for analysing signatures both within a single Capture Session and over time (multiple Sessions). Secondly, the physical characteristics of signatures which result in the most repeatable performance for each feature are also analyzed. These results have implications for biometric signature verification systems and the document forensic field in that it gives an indication as to the stability of features leading potentially to improved performance and the types of features that should be analyzed given particular characteristics of the signature under investigation.