One fundamental difference between causal methods and epipolar and trifocal techniques is that only small baselines are available as the camera or objects are displaced incrementally. Thus, techniques that are sensitive in this small displacement range will exhibit numerical problems. The reality is that there is little to no SfM information on a frame to frame basis, especially given the possible noise on the image features. This is a difficulty for almost any two-frame SfM algorithm. Thus, one must consider integrating (in batch or recursive form) the information over the sequence.
An important advantage arises in small baseline situations, though. The correspondence problem and feature tracking become easier. The proximity of features over adjacent frames can be used to guide correspondence and most feature detection techniques will exhibit more consistent behavior as images change slowly.