Limits of 2D Simulation in Facial Harmonization

Image simulation is a useful and honest tool when you know what it is. It is a transformed photo, in two dimensions, meant to illustrate one possible direction. The problem arises when more is expected of it than it delivers: volume precision, result prediction, a portrait of the face in motion. The image is convincing, and so it must come with a clear map of its limits.

This page lists what 2D simulation does not represent, why each limit matters in HOF practice and how to turn that knowledge into an honest talk with the patient. Knowing the limits is what makes the tool trustworthy.

What is 2D simulation?

It is the generation of a flat image from a photo, with changes in chosen areas. It has a single view (the frontal one, in Harmy's case), with no depth information, no camera moving around the face and no dynamics. The result is an illustrative simulation: a visual representation of the plan's idea, not a measure of what will happen.

The contrast with the actual result is covered in why the actual result differs from the simulation. AI generation technology is discussed in AI simulation: what it changes and what it does not.

What does a 2D image not represent?

The limits can be grouped into eight items.

Several of these limits have their own pages: simulation does not show movement, swelling gets in the way of comparing results and lighting for the planning photo.

Why does the lack of profile matter?

Many HOF requests depend on projection, which can only be seen from the side: the nose, lips, chin, forehead. A frontal photo does not contain that information, and a frontal simulation may show a "bigger" lip without saying how far it projects, or a "more present" chin without showing the effect on the profile.

The practical result is that, for lips, chin and nose, the frontal simulation illustrates apparent width, height and contour, but not projection. Those aspects must be assessed on a side photo, in person and, when needed, with specific tools. Reading the profile is covered in aesthetic profile: lips, nose and chin and profile facial analysis: checklist.

What about movement and expression?

The face lives in motion, and the result of a volume behaves differently at rest and in expression. A lip that looks perfect when still may have a different shape when smiling. A cheekbone projected at rest may change the nasolabial fold when smiling. The simulation does not capture these changes.

That is why the image is a snapshot: it says how the plan may look at rest, not how the face will react in use. Dynamic assessment is covered in dynamic facial analysis in expression and face at rest vs in motion.

Can AI get the image wrong?

It can. Image generation models sometimes alter details they should not: they change a feature, soften an asymmetry, correct a characteristic. This kind of error is described as AI image hallucination. Even with rules that limit changes to the marked areas only, the professional's check is essential.

Points to watch when reviewing the image:

• did the image change only what was marked?
• were characteristic features (asymmetries, moles, eye shape) preserved?
• did the expression, hair and background stay the same?
• does the volume look consistent with the planned ml?
• are there visual artifacts, odd edges or blurred areas?

If something does not match, the image should not be shown as it is. The conversation about simulation errors appears in simulation with a bad photo and the patient does not like the simulation.

How do you turn the limits into an honest talk?

Knowing the limit is the first step; saying it is the second. A simple script covers the essentials.

This short script sums up the main limits without taking away the tool's value. After it, the image becomes what it should be: an aid to the conversation. The full design of the conversation is in how to explain that the simulation is not a guarantee.

When does 2D simulation really help?

Despite the limits, the tool has legitimate and valuable uses:

1. Communicate direction and proportion. The patient sees where the plan's focus is.
2. Calibrate expectations. Seeing a discreet version helps drop the idea of excess.
3. Compare intensities. Subtle, Moderate and Pronounced side by side.
4. Align priorities. The image shows which areas change the appearance most.
5. Document the agreement. The saved image shows what was presented.

The discussion on when it helps and when it gets in the way is in when simulation helps the plan and when simulation gets in the way of the plan.

What precautions should you take when using it?

The broader ethics are in ethics of result simulation, and record keeping in recording the simulation in the patient record.

Quiz: limits of simulation

Summary

• 2D simulation is a flat, static image made from a frontal photo.
• It does not represent depth, profile, movement, texture, swelling, time or individual response.
• Projection decisions call for a side photo and an in-person assessment.
• AI can get details wrong: check the image before showing it.
• Say the limits out loud and record them in the patient record.
• Well used, it helps communicate direction and calibrate expectations.