Building on the observation that finger orientation is an inherent part of human’s interaction in the real world, exploiting finger orientation for multi-touch tabletop interaction would facilitate more natural interaction techniques. We motivate this by means of examples where the finger orientation improves or enriches interaction. Afterwards, we present a simple and fast approach to detect the finger orientation reliably for multi-touch tabletop interaction. The steps involved are computationally cheap and therefore suit the needs of tracking software operating under time-critical conditions. We show that the presented approach enables the detection of finger orientation also for fingers that touch the tabletop surface only slightly. Further, recognition rates on real data gained from the camera within a multi-touch tabletop are presented in order to give a measure for the precision and reliability of the presented approach.