Following the Memorandum OF Understanding between the CHINA NATIONAL SPACE ADMINISTRATION (CNSA) and the NETHERLANDS SPACE OFFICE (NSO) concerning the cooperation on the Chang’E-4 mission, the Netherlands-China Low frequency Explorer (NCLE) payload is currently being developed by the joint Sino-Dutch teams. NCLE is regarded as a pathfinder mission for a future space-based low-frequency radio interferometer which aims at investigating the evolution and formation of the structures in the Dark Ages and Cosmic Dawn. Here we propose for two PhD positions to prepare for and to execute the science exploitation of the NCLE payload. The PhDs will become part of the NCLE team, collaborating with both scientist and engineers on the Dutch and Chinese side. We propose that in phase I of the project the PhDs take part in the commissioning of the NCLE instrument, while one PhD has the focus on the calibration of the instrument and the other focusses more on the design, implementation and calibration of the data pipeline. In phase II we identify constraining the 21-cm line emission global Dark Ages and Cosmic Dawn signal as the major science case, while the study of the emission from the Sun and the large planets is considered science that can easily be addressed by NCLE during the 4 year PhD tracks. In addition, a number of ancillary science cases are identified, ranging from the production of a low-frequency radio sky map to the study of the Lunar ionosphere. Finally, in collaboration with technical teams at the Chinese and Dutch side, we propose that an important outcome of the 2 PhD tracks will be a future concept for a space-based low-frequency radio facility. The work proposed here supports this unique Sino-Dutch project and helps preparing for the science exploitation of NCLE well in time for the launch in May 2018.

Promo and trailer production are expensive and labor-intensive tasks in media industry. Recently, content-based video summarization has found its way to the field of computer vision and deep learning that can greatly reduce the human effort for this task. Most effective methods in this regard are relying on deep learning approaches, i.e., learning effective video representations directly from raw video frames, without hand crafting them for the summarization task. Deep learning frameworks are known to be data hungry, which require an immersive annotated data for the training procedure. Collecting annotated video for training deep networks is yet another labor-intensive task that should be limited for cheap automatic promo production. The common solution to avoid the annotation costs is to use the pretrained network on existing labeled datasets. Nevertheless, the performance of the pre-trained networks deteriorates gracefully when the properties of the test videos drifts away from that of the training videos. In this project, we aim at quantifying/modeling the visual domain disparity for the video summarization task by means of real data that is provided by our private partner (RTL). This can be effectively done by evaluating the existing deep learning approaches, that are trained and tested on public video datasets, directly on RTL video collections. Moreover, the evaluation of video summarization task is hardly objective due to its complex nature (human reasoning is required). We benefit from the expertise of our industrial partner to develop a framework for instrumental measure of the quality of summarized video content.