There are some standardized methods for typing with VI:
N vs S: N eyes are usually not focused on their surroundings -- they have direction but not slant. S eyes are about always slanted toward one object in particular that they are studying at that moment.
T vs F: F expressions look more emotional, like those of a dog. T expressions seem more poised and calculated, like a cat's. F also has a tender physical frame, where T has a very structured frame and a muscular build to match.
I vs E: difficult to determine without full motion video, however E is more likely to look directly at their interlocuter than I, who will usually demur and cast their glance to the side as they speak. (away from the eyes)
J vs P: P has an angular face that will appear smaller than J's when body mass is similar; J has strong, rigid features. P rolls with the punches, J is built to absorb the punch full on. Their builds reflect evolutionary coherence with their styles of interaction.