https://en.wikipedia.org/wiki/Group_of_pictures
Most important part:
The I frames contain the full image and do not require any additional information to reconstruct it. Typically, encoders use GOP structures that cause each I frame to be a "clean random access point," such that decoding can start cleanly on an I frame and any errors within the GOP structure are corrected after processing a correct I frame.
P and B frames contain motion-compensated difference information relative to previously decoded pictures.
So, it is very important not to break a GOP when you cut the movie because if an I frame is missing, the next P/B frames will contain relative data to an unknown reference frame (the I frame). This is why you have pixelation for short periods, until a new intact GOP is reached.