River is one of the most significant factors in driving the formation and evolution of landforms as well as one of the most important freshwater resources on the Earth. River characteristics, including water extent, water level/water depth, river discharge, water quality, and ice coverage, are vital to the dynamic monitoring of water resources and protection of eco-hydrological systems. Traditional methods of acquiring river characteristics are to use in situ data that were collected on hydrological gauges, which costs large amount of financial and material resources. It is urgent to develop a way of supplementing in situ data of rivers quickly and accurately for the increasing demand of river information. With the ability of omnidirectional and multi-temporal Earth observation, remote sensing has greatly improved the efficiency of acquiring river characteristics. It has been applied broadly in multi-scale river monitoring and hydrological simulation in ungauged basins. Therefore, it is necessary to summarize recent progresses in the field of remote sensing based river characteristics inversion, in order to further promote the application of remote sensing data and methods in this field. This article, therefore, focusing on remote sensing of river characteristics, summarizes recent progresses systematically on the extraction of water extent, inversion of water level/water depth, estimation of river discharge, and monitoring of water quality and ice coverage. Advantages and disadvantages of applying optical and microwave remote sensors for obtaining water extent and water level/depth are discussed in detail. The advanced data, specific methods, and related emerging technologies in this field are discussed and the following conclusions are made: 1) Newly available remotely sensed data have been making creative breakthroughs in spatial resolution, temporal resolution, and spectral resolution, which dramatically enrich data sources for river studies. 2) Most optical images still face the challenge of mixed pixels, while the application of SAR images is suffering from difficulties in developing complex processing algorithms. Meanwhile, accurate extractions of narrow and non-open waters need further research in the future. 3) The development of big data and cloud computing technologies provide excellent means for monitoring river characteristics at large spatial scales and long temporal scales, with both high spatial and high temporal resolutions.