Why the Camera Adds 10 Pounds: Seeing Ourselves In Pictures

A week ago, a short TED talk by Duncan Davidson called “Why do we hate seeing photos of ourselves” went viral in the blogosphere. While I agree with Duncan’s main premise that part of the issue is that we are used to seeing a mirror image of ourselves, I think it goes deeper.

Some of the responses to the post touch on more issues, such as the image being static and missing “micro-expressions.” I agree with those, too. But I think the static image and micro expressions are more related to the retouching that we do. In real-life interactions we don’t notice little zits, scars, pores, etc. as the face is moving and our eyes are moving around the face of the person we are talking to. In the captured still all of these things stand out to us in a big print, but I don’t think we see that much detail in the small image on the back of the camera that is referenced in the video.

I think there is another issue that wasn’t brought up in the talk or the comments. That is the camera-to-subject distance. As photographers, we are constantly instructed to be at least 36 inches from our subject. Many instructors tell us to be 5 feet or 6 feet away. This is supposed to give a “better” or more pleasing perspective. And that may well be so.

In many of our social interactions we are probably around 2 to 3 feet or more away from the people we are talking to. At least in the US culture. Yes, there are cultures where it is more common to be only inches away from the people you are talking to. But I don’t have experience photographing in those places. I’d be interested in hearing from people who do usually stand “in each others’ face” about how subjects react to seeing their photographs. In the case of most of the subjects I work with, we are used to seeing other people from a bit of a distance.

But how do we see ourselves? I usually see myself in the bathroom mirror. How about you? Today I measured the distance from my eye to the mirror when I’m standing at the sink. That was 21″. I checked with my wife, Kim, and she said she is usually looking in a makeup mirror. We went and measured and found she was 9″ from the mirror.

When it comes to perspective in photography, the closer you are to your subject, the more narrow their face will appear. When you are just inches away from your subject their nose is relatively much closer to you than their eyes and ears. The ears seem farther away, making them smaller and making the face seem narrower.

So, my additional premise on this is that we all seem to look heavier to ourselves in photographs that are taken from greater distances. Other people will think the photo looks good because they are used to seeing us from those greater distances. And in a mirror we are always looking at ourselves at eye level. The woman in Duncan’s photo was also photographed from above. She probably has never seen herself in a mirror from that view.

A few weeks ago a question came up on Quora that I was compelled to reply to. It was about why the camera “adds 10 lbs.” to people. You can read my response there.

Here are some photos to help illustrate. All of them were taken with the same lens (Canon 24-105mm) set to 24mm at different distances, from 18″ to 60″ (measured from the mannequin’s right eye to the focal plane mark on the top of the camera). They were then cropped so that the face is approximately the same size in each frame…

Here are the 18″ and 60″ images side-by-side…

I think you will agree that in the closer photos taken at 18″, 21″, and 30″ make our subject’s face appear thinner. This is how you would see yourself in the mirror. As we then move back to 48″ and 60″ away the face starts to flatten out and look a little heavier. To someone really concerned about their appearance, it might see more than a little heavier. That extra 10 pounds that we get from being further away.

As noted, all the above images were taken with a 24mm lens. This is to show that it isn’t the lens that does this, but is instead the camera-to-subject distance. Don’t blame the lens!

You may be asking yourself about the difference in the appearance of the backgrounds. The lighting is the same (a Westcott Apollo Orb on camera left with a home-made crinkled aluminum foil covered reflector on the right). What is different is that from the greater distance we are only seeing the middle of the backdrop right behind the mannequin’s head. When in closer the field of view is wider and we are seeing the edges of the backdrop and see the falloff of the light. Here is a quick diagram to help explain the field of view…

To help illustrate that this is not lens focal length related, this next image was taken by zooming the lens to 105mm without moving the camera after the 60″ image from above. You will see that the perspective is exactly the same as in the 24mm at 60″ image above.

Here they are side-by-side and then overlayed on each other in Photoshop in Difference mode.

Notice that only the very edges of objects are highlighted as different between the two images. That is most likely because the depth of field is different between the two and the quality of the 105mm image is better because it isn’t cropped in so much (losing resolution). But the look of the face is the same in the photos with both the 24mm and 105mm lenses.

Here is a little experiment for you the next time you are standing in front of the bathroom mirror. Lean in real close to the mirror and study how your face looks. Then slowly pull back away from the mirror while carefully observing the look of your face. Notice that the perspective flattens out as you back away, which gives the impression of being slightly wider, which we translate into appearing those ten pounds heavier.

And one more question… Do you like photos you take of yourself with your phone camera? What is the distance you can hold your phone away from your face to take those photos?

About the author: John Cornicello is a photographer based in Seattle, Washington. Visit his website here. This post was originally published here.

Image credit: Wavy Self Portrait by Robert Hruzek, Mirrors by shareski

  • Greg Planchuelo

    Very nice and interesting post.

  • Dana

    Very interesting, never realised that! Thank you for the post and info! :)

  • Kate Hailey

    Great Post! John, I love the break down of the setup and the step by step here, great way to explain the concept to all of us visual folks! ;)

  • BArb

    Fascinating. Thank you for explaining this.

  • Noxonomus

    The view you see of your self in a mirror from 21″ would look like a photo form 42″ away (halfway between the 36″ and 48″ photos) and not really much like the photos from 18″ 21″ or 30″ inches. In the case of your wife however the 18″ photo really is like her view in the makeup mirror unless it is a curved magnifying mirror.

  • Ze Courlis

    I suppose it’s one of the reasons why fashion designer always choose very thin models to show their collections …

  • madeline nel

    Thankyou for sharing this information with us John!

  • Marcia

    When I take a photo of myself with my iPhone camera, my nose is distorted, it looks bigger than it does just looking in a mirror.

  • Greg Downing

    It has been known for some time in stereoscopic photography circles that 3d pictures take 10 pounds off people. When you see someone in 2d they are wide, in 3d when you can see their width in proportion to their depth they look thinner. When you are using a mirror you are looking at yourself in 3D. This may compound the effect that you are describing.

  • Earl Newton

    Wouldn’t there be less distortion in a longer lens? (i.e., 50mm, 100mm)

  • Guest

    Seriously? You shoot with a 24mm and not expect wide angle distortion to affect the shape of their faces? It’s a good sharing but somewhere along the way you’re getting something wrong about lens physics.

  • Akshay Chauhan

    This is exactly my thoughts were when I saw the TED video. Not so surprisingly the people who actually take weird portraits won’t realize that its bad.

  • Akshay Chauhan

    It also depends on stereobase/interocular distance (the distance between the L and R camera). If its more than normal then person/object would be dwarfed or if its less he will become larger.

  • TheBob

    I think there is more going on in the brain’s interpretive “software” than in any physics of lenses. I think our brains take in the information we want to take in in person, because we know its a person, and we take in the stark reality of a photograph in 2D because our brains se an image, so to speak, rather than a real person.

  • John Cornicello

    Yes, seriously.

    When we talk about “distortion” in lenses we have to be careful. To a lens designer distortion is a specific condition where the corners of an image pull in or pull out. We commonly call this barrel or pincushion. It is independent of focal length. It has to do with the lens design and where the aperture is placed within the groups of elements in the lens. In some lenses (zooms) you may have pincushion distortion at some focal lengths and barrel distortion at others.

    Next we have perspective, which we’ve been talking abou in this blog. Again, independent of focal length. Perspective is determined by camera to subject distance. The focal length of the lens just determines the field of view and magnification. When used at the same camera to subect distance a wide angle, normal, or long lens will all have the same perspective in the portion of the scene that is common to all three lenses. The wide angle lens will have a wider field of view and will include more of the scene. But it won’t change the perspective unless you move the camera.

    Then there is what is often called wide-angle distortion. That is not what we see with someone’s nose being too large from being too close. Instead it is more noticeable at the edges of the image. Part of that is just physics. If you take a wide angle photo you are taking an infinitely wide scene and compressing it down to a fixed width of the film or sensor. That does stretch things at the edges of the frame. And there is also the issue where if you take a photo of a group of people or objects with a short lens those subjects on the edges seem to get stretched out compared to the subjects in the middle of the frame. This is actually also a function of camera-to subject distance. When in close to a group the camera/lens is seeing the middle subjects straight on. But it is seeing the subjects on the edges from a more severe angle and is seeing both the front and the side of those subjects, leading to that stretched out look. You can somewhat compensate for this by having people on the edges of a group turn in slightly towards the group. But you still run into the issue of compressing the wide scene onto the film/sensor.

  • Leo Sands

    A question.
    The face looks slimmer means the face is elongated vertically.
    That means that the ratio of vertical dimension to horizontal dimension of the face increases as the distance between camera and the object decreases.
    The photos seems to be taken in portrait orientation. Imagine that the model is laying on the ground and one takes the photos again with the same settings and in portrait orientation, the face of the model will be fatter in photo with closer distance from the camera. The result will be reversed.
    How about using a circle as a test?
    Is this effect related to the science of optics?

  • LR

    And actresses aren’t even thinner like that anymore these days. They’re curvy like most average women as well as athletic.