The aim of this study was to test the accuracy and the reliability of 3-dimensional angular measurements of an instrumented knee brace compared to the gold standard navigation system optoelectronic cameras-based. Thirteen cadaveric bodies were used to calculate kinematic knee parameters. Three exercises were performed, 100° flexion (FLEX100), internal/external rotation with a 30° flexion (ROT30), and the pivot shift test (PS). The reliability was excellent with an ICC (95%) > 0.90. The agreement between the two systems showed excellent correlation in the F/E axis for FLEX100 and PS (σ (Flex/Ext) > 0.95) and strong correlation in the I/E axis for ROT30 (σ (Rot Int/Ext) > 0.939). The root mean square error (RMSE) was under 5° for all exercises considering the soft tissue artifact (STA) for the F/E axis. Consequently, the instrumented knee brace exposed high reliability and accuracy which could end up on clinical interpretations thanks to the previous measures.