Traditionally, arboreal rainforest mammals have been inventoried using ground-based survey techniques. However, given the success of camera traps in detecting secretive terrestrial rainforest mammals, camera trapping could also be a valuable tool for inventorying arboreal species. Here we assess, for the first time, the effectiveness of arboreal camera traps for inventorying arboreal rainforest mammals and compare the results with those from other methodologies. We do so in one of the world's most biodiverse conservation areas, the Manu Biosphere Reserve, Peru. We accumulated 1201 records of 24 arboreal mammal species. Eighteen species were detected by arboreal cameras, seven by diurnal line transects, six by nocturnal transects and eighteen through incidental observations. Six species were only detected using arboreal camera traps. Comparing arboreal camera traps with traditional ground-based techniques suggests camera traps are an effective tool for inventorying arboreal rainforest mammal communities. They also detected more cryptic species compared with other methodologies. Daily detection frequency patterns were found to differ between ground-based techniques and arboreal cameras. A cost-effort analysis indicated that despite greater upfront costs in equipment and training for arboreal camera trapping, when accounting for the additional survey hours required to provide similar numbers of records using ground-based methods, overall costs were similar. Our work demonstrates that arboreal camera trapping is likely to be a powerful technique for inventorying canopy mammals. The method has considerable potential for the study of charismatic and threatened arboreal mammal species that may otherwise remain largely unknown and could quietly disappear from the world's tropical forests.