The collection of high-resolution data is helping researchers better understand form, process, and change in river systems and, especially, stream restoration projects. In the last 10 years it has become apparent that there is no “one-size-fits-all” approach to collecting high-resolution data for fluvial studies. The approach we demonstrate here is a self-contained pole aerial photography (PAP) system capable of collecting data for directly georeferenced structure from motion photogrammetry. PAP can produce higher-spatial resolution data than remotely piloted aerial system collected data and is one option where RPAS are restricted, like in parks and protected places. Another advantage of PAP is that it makes 3D data collection possible in parts of rivers, like under riparian canopies, that can elude capture with RPAS methods. Direct georeferencing removes the necessity for ground control points, which can greatly decrease the amount of time needed for a survey. The system that we developed combines a low-cost, dual-frequency GPS receiver capable of Real-time and Post-Processed Kinematic surveying with an off-the-shelf digital SLR camera, an inertial measurement unit, and 3D printed mounts and housings. The open-source control/survey software runs on an inexpensive Raspberry Pi computer with a 7-inch (18 cm) touch screen display. We highlight the accuracy of the system along with the high spatial resolution 3D data, ortho imagery, as well as other data that can be derived from these datasets such as sediment size measurements.