Berta Bescos, Jose Neira, Roland Siegwart, and Cesar Cadena IEEE International Conference on Robotics and Automation (ICRA) 2019 In this paper we present an end-to-end deep learning framework to turn images that show dynamic content, such as vehicles or pedestrians, into realistic static frames. This objective encounters two main challenges: detecting all the dynamic objects, and inpainting the static occluded background with plausible imagery. The former challenge is addressed by the use of a convolutional network that learns a multiclass semantic segmentation of the image. The second problem is approached with a conditional generative adversarial model that, taking as input...
Technical Report 2 Explanation of the work carried out by the beneficiaries and Overview of the progress. pdf
Technical Report 1 Explanation of the work carried out by the beneficiaries and Overview of the progress. pdf
Press video This deliverable provides the UP-Drive press video. The video itself is placed at the landing page of the project webpage at https://www.up-drive.ethz.ch – it is the bottom video. pdf
Mid-term demonstration This document reports on UP-Drive mid-term demonstration which marks the successful completion of the 1st development and integration cycle pdf
Initial dissemination report The UP-DRIVE dissemination report contains detailed descriptions of the strategy and the means by which UP-DRIVE is communicated. pdf
Brochure, newsletter The Newsletter is dedicated to presentation information about the UP-Drive project. pdf
Project Web-page This deliverable corresponds to Task 9.2: Set-up of fully functional project web site. In the current stage the web site offers detailed information about the project and of the partners in UP-Drive consortium. The web site will be updated continuously to communicate to the public any project related news. pdf
Evaluation report on integration process and results of first development cycle This report explains the integration process and highlights the results of the first of the two development cycles. pdf
Integration and test tools and processes This deliverable describes the integration tools and the processes established by the consortium. The choice of the tools and the processes is based on best practices from previous collaborative robotic projects. pdf