Zhihan Lv

TL;DR: This paper proposes a hybrid framework by interleaving a Convolutional Neural Network and a traditional optical flow energy, which enables the capability of modeling and solving both the blind deconvolution and the optical flow estimation problems simultaneously.

TL;DR: A cloud service platform is proposed, that targets wise transportation in order to carry out unified management and mining analysis of the huge number of the multivariate and heterogeneous dynamic transportation information, provides real-time Transportation information, increases the utilization efficiency of transportation, promotes transportation management and service level of travel information, and provides decision support of transportation management by virtual reality as visual.

TL;DR: A simple and efficient feature tracking method based on the standard nearest neighbor search with the best performance in both speed and accuracy, and a geometry-constraint approach to remove outliers from the initial matches for boosting the matching precision.

TL;DR: IoT monitoring of geospatial images, videos, and associated optimization and control algorithms should be a research priority towards expanding insights for soil abiotic CO2 uptake and a better understanding of how the uptake happens in arid region.

TL;DR: Voluntary computing can be applied to the processing of big data, which has a good reference significance for the distributed analysis of large-scale complex networks.