Systems and methods for endoscopic procedures employ a first technique to ensure initial correct alignment of an endoscope (100) with a desired target (10). A reference image (51) is then acquired from an imaging arrangement associated with the endoscope. During a subsequent stage of the procedure, tracking of the endoscope position relative to the target is performed partially or entirely by image-based tracking by comparing features in real-time video image (52) produced by imaging arrangement with features in the reference image (51). The feature comparison may be performed visually by a user, or may be automated to offer more specific corrective suggestions to the user.
