Project MARVEL will create and publicly share with the academics, industrial community, and smart... more Project MARVEL will create and publicly share with the academics, industrial community, and smart cities, a data pool of experimental multimodal audio-visual data and will showcase the use of the data and its processing across various pilots. This report documents the process for the collection and analysis of the experimental data. The use cases and AI tasks required to process the audio-visual data and enable the implementation of the pilots are first described. This knowledge determines how and where the audio-visual data is collected and annotated. The various devices, namely microphones and cameras, and their deployment are described next, followed by software tools that will be used in the data annotation task, that will be carried out according to what is required for training the AI models. The data is analysed to determine which parts constitute personal data, followed by a discussion on the appropriate data anonymisation techniques, which should ensure the sharing of GDPR ...
When dealing with multi-class classification problems, it is common practice to build a model con... more When dealing with multi-class classification problems, it is common practice to build a model consisting of a series of binary classifiers using a learning paradigm which dictates how the classifiers are built and combined to discriminate between the individual classes. As new data enters the system and the model needs updating, these models would often need to be retrained from scratch. This work proposes three learning paradigms which allow trained models to be updated without the need of retraining from scratch. A comparative analysis is performed to evaluate them against a baseline. Results show that the proposed paradigms are faster than the baseline at updating, with two of them being faster at training from scratch as well, especially on larger datasets, while retaining a comparable classification performance.
In this paper we describe $Vja\dot{g}\dot{g}$, a battery-aware journey detection algorithm that e... more In this paper we describe $Vja\dot{g}\dot{g}$, a battery-aware journey detection algorithm that executes on the mobile device. The algorithm can be embedded in the client app of the transport service provider or in a general purpose mobility data collector. The thick client setup allows the customer/participant to select which journeys are transferred to the server, keeping customers in control of their personal data and encouraging user uptake. The algorithm is tested in the field and optimised for both accuracy in registering complete journeys and battery power consumption. Typically the algorithm can run for a full day without the need of recharging and more than 88% of journeys are correctly detected from origin to destination, whilst 12% would be missing part of the journey.
Automatic description generation from natural images is a challenging problem that has recently r... more Automatic description generation from natural images is a challenging problem that has recently received a large amount of interest from the computer vision and natural language processing communities. In this survey, we classify the existing approaches based on how they conceptualize this problem, viz., models that cast description as either generation problem or as a retrieval problem over a visual or multimodal representational space. We provide a detailed review of existing models, highlighting their advantages and disadvantages. Moreover, we give an overview of the benchmark image datasets and the evaluation measures that have been developed to assess the quality of machine-generated image descriptions. Finally we extrapolate future directions in the area of automatic image description generation.
2009 Second International Conference on Advances in Circuits, Electronics and Micro-electronics, 2009
Page 1. An Efficient Algorithm for the Control of Reconfigurable Pixel Microstrip Antennas Adrian... more Page 1. An Efficient Algorithm for the Control of Reconfigurable Pixel Microstrip Antennas Adrian Muscat Dept of Communications and Computer Engineering University of Malta, Malta Email: adrian.muscat@um.edu.mt Joseph ...
This paper describes a method using the planar circuit technique together with the Generalized Ca... more This paper describes a method using the planar circuit technique together with the Generalized Cavity Model (GCM) and segmentation/desegmentation methods to calculate the resonant frequency and input reflection coefficient of a pixel patch antenna. Tree and graph representations are developed to represent the pixel patch antenna and to reduce the antenna to a network of connected ports. The segmentation technique
IEE National Conference on Antennas and Propagation, 1999
A full-wave 3-D FDTD model is used to analyse a finite array of aperture coupled microstrip patch... more A full-wave 3-D FDTD model is used to analyse a finite array of aperture coupled microstrip patch antennas. The mutual coupling between elements and input impedance are calculated and compared to measured data.
Project MARVEL will create and publicly share with the academics, industrial community, and smart... more Project MARVEL will create and publicly share with the academics, industrial community, and smart cities, a data pool of experimental multimodal audio-visual data and will showcase the use of the data and its processing across various pilots. This report documents the process for the collection and analysis of the experimental data. The use cases and AI tasks required to process the audio-visual data and enable the implementation of the pilots are first described. This knowledge determines how and where the audio-visual data is collected and annotated. The various devices, namely microphones and cameras, and their deployment are described next, followed by software tools that will be used in the data annotation task, that will be carried out according to what is required for training the AI models. The data is analysed to determine which parts constitute personal data, followed by a discussion on the appropriate data anonymisation techniques, which should ensure the sharing of GDPR ...
When dealing with multi-class classification problems, it is common practice to build a model con... more When dealing with multi-class classification problems, it is common practice to build a model consisting of a series of binary classifiers using a learning paradigm which dictates how the classifiers are built and combined to discriminate between the individual classes. As new data enters the system and the model needs updating, these models would often need to be retrained from scratch. This work proposes three learning paradigms which allow trained models to be updated without the need of retraining from scratch. A comparative analysis is performed to evaluate them against a baseline. Results show that the proposed paradigms are faster than the baseline at updating, with two of them being faster at training from scratch as well, especially on larger datasets, while retaining a comparable classification performance.
In this paper we describe $Vja\dot{g}\dot{g}$, a battery-aware journey detection algorithm that e... more In this paper we describe $Vja\dot{g}\dot{g}$, a battery-aware journey detection algorithm that executes on the mobile device. The algorithm can be embedded in the client app of the transport service provider or in a general purpose mobility data collector. The thick client setup allows the customer/participant to select which journeys are transferred to the server, keeping customers in control of their personal data and encouraging user uptake. The algorithm is tested in the field and optimised for both accuracy in registering complete journeys and battery power consumption. Typically the algorithm can run for a full day without the need of recharging and more than 88% of journeys are correctly detected from origin to destination, whilst 12% would be missing part of the journey.
Automatic description generation from natural images is a challenging problem that has recently r... more Automatic description generation from natural images is a challenging problem that has recently received a large amount of interest from the computer vision and natural language processing communities. In this survey, we classify the existing approaches based on how they conceptualize this problem, viz., models that cast description as either generation problem or as a retrieval problem over a visual or multimodal representational space. We provide a detailed review of existing models, highlighting their advantages and disadvantages. Moreover, we give an overview of the benchmark image datasets and the evaluation measures that have been developed to assess the quality of machine-generated image descriptions. Finally we extrapolate future directions in the area of automatic image description generation.
2009 Second International Conference on Advances in Circuits, Electronics and Micro-electronics, 2009
Page 1. An Efficient Algorithm for the Control of Reconfigurable Pixel Microstrip Antennas Adrian... more Page 1. An Efficient Algorithm for the Control of Reconfigurable Pixel Microstrip Antennas Adrian Muscat Dept of Communications and Computer Engineering University of Malta, Malta Email: adrian.muscat@um.edu.mt Joseph ...
This paper describes a method using the planar circuit technique together with the Generalized Ca... more This paper describes a method using the planar circuit technique together with the Generalized Cavity Model (GCM) and segmentation/desegmentation methods to calculate the resonant frequency and input reflection coefficient of a pixel patch antenna. Tree and graph representations are developed to represent the pixel patch antenna and to reduce the antenna to a network of connected ports. The segmentation technique
IEE National Conference on Antennas and Propagation, 1999
A full-wave 3-D FDTD model is used to analyse a finite array of aperture coupled microstrip patch... more A full-wave 3-D FDTD model is used to analyse a finite array of aperture coupled microstrip patch antennas. The mutual coupling between elements and input impedance are calculated and compared to measured data.
Uploads
Papers by Adrian Muscat