Search

In this paper, we propose a method for coarse camera pose computation which is robust to viewing conditions and does not require a detailed model of the scene. This method meets the growing need of easy deployment of robotics or augmented reality applications in any environments, especially those for which no accurate 3D model nor huge amount of ground truth data are available. It exploits the ability of deep learning techniques to reliably detect objects regardless of viewing conditions. Previous works have also shown that abstracting the geometry of a scene of objects by an ellipsoid cloud allows to compute the camera pose accurately enough for various application needs. Though promising, these approaches use the ellipses fitted to the detection bounding boxes as an approximation of the imaged objects. In this paper, we go one step further and propose a learning-based method which detects improved elliptic approximations of objects which are coherent with the 3D ellipsoid in terms of perspective projection. Experiments prove that the accuracy of the computed pose significantly increases thanks to our method and is more robust to the variability of the boundaries of the detection boxes. This is achieved with very little effort in terms of training data acquisition -- a few hundred calibrated images of which only three need manual object annotation. Code and models are released at https://github.com/zinsmatt/3D-Aware-Ellipses-for-Visual-Localization., Comment: Presented at 3DV 2020. Code and models released at https://github.com/zinsmatt/3D-Aware-Ellipses-for-Visual-Localization

The digital battle has become the most important factor of economic and social development. In fact, there is no doubt on the fact that establishing E-government systems can improve the quality of government services and enhance the citizens’ life. However, E-government projects cannot be successful if they are not adapted to the specific cultural and social needs of the final users. Taking into consideration the high illiteracy rate within the Moroccan population, we conducted our research in order to design an E-government portal for the old city of Fez that is accessible by the majority of the citizens especially those who are illiterate. The objective of this research is exploring the different architectures, scenarios and technologies for designing an E-government portal that is usable and accessible by all Fez citizens whether they are literate or illiterate. Hence, the concrete output of this project is a usable graphical user interface accessible to both literate and illiterate people in order for them to interact with the static content and to use the online transactions. In the process of achieving our goal, we had to discover and create design principles and to test them with real users. To achieve our goal we had to pay particular attention to the following points: • Considering the two existing types of illiteracy: literal illiteracy and digital illiteracy. • Designing the GUI so that it is not dedicated mainly to illiterate people: literate people too should be able to use the e-government system without frustration. • Since it is the first portal targeting illiterate users in Morocco, we were interested also in building conventions on the type of symbols to be used. In fact, the main two assumptions behind this project are: • Effective electronic service delivery has to enable citizens to access government services in a new and more effective way than the traditional ways. • Government services have to be presented to citizens in a usable format and in a human centred manner. In that way, we can express the goals that are aimed to be achieved by this project as being: • Improving knowledge about how illiterate citizens in Morocco can interact with E- government services. • Developing a methodology and architecture for e-government application development that fulfils the needs of Arabic developing countries. • Demonstrating the ability of Moroccan illiterate citizens to interact with e-government portals that uses new technologies. This paper describes and justifies the chosen methodology, architecture and tools that were applied in order to develop Fez e-government portal’s graphical user interface. The results of the designed graphical user interface’ s testing are presented and discussed in order to assess the project’s success., Revue Marocaine des Sciences de Management, No 7 (2017)