Eyebrows and Face Recognition

Do you recognize these two people? In both photographs, the eyebrows have been removed. 

Here are the photos of the same faces. Is it easier to recognize them this way? This time the eyes have been digitally removed instead of the eyebrows. (Hint: one is a politician, and the other an actor) 

Richard Nixon and Winona Ryder

Scientists have done facial recognition experiments where subjects were presented with many faces altered to have either the eyebrows or the eyes removed. It turns out that subjects perform better on faces with no eyes, compared to faces with no eyebrows. 

As the authors put it, "The absence of eyebrows in familiar faces leads to a very large and significant disruption in recognition performance."

This came as a surprise to me, since I have always assumed that the eyes were the most important elements to help us recognize and remember a face, with the mouth being perhaps second most important.

Anselm van Hulle, 1649. Anna Margareta

It's remarkable that humans of both sexes have these patches of hair on our faces, compared to primates who generally have more facial hair. The muscles controlling their movement are sophisticated and largely unconscious. We express much about our emotional state to others, even at long distances away. This central role as a social signaler may be related to why eyebrows are also so important for recognition.

The authors of the paper note that:

"During the 18th century, in fact, in Western Europe full eyebrows were considered so essential to facial beauty that some upper-class women and courtiers would affix mouse hide to their foreheads. The perceived importance of the eyebrows for enhancing beauty has not waned to this day. Currently, it is relatively common cosmetic practice to use tweezers or depilatories to narrow and accentuate the arch of the eyebrows, as well as to remove any hair at the bridge of the nose. Cosmetics may also be used to alter the color (especially the darkness) and exaggerate the shape and length of the eyebrows."

----
J. Sadr, I. Jarudi, and P. Sinha, B, The role of eyebrows in face recognition, [Perception, vol. 32, pp. 285–293, 2003.

Lighting a Model with Two Sources

Sir John Leighton, the Director-General of the Scottish National Gallery, served as both painter's model and keynote speaker at the Portrait Society's annual conference yesterday.

Sir John Leighton by James Gurney, black and white gouache, 3 x 3 inches

He was on the grand ballroom stage posing for a demo by Michael Shane Neal. I was far back in the audience watching the demo, looking at a video image projected on a big screen. Above is a 30-minute gouache sketch I did from my seat.

Mr. Neal lit him with a two-source lighting scheme inspired by Anders Zorn (Swedish, 1860-1920). The lighting scheme produces a shadow core in the center of the form and often puts the eyes in shadow.

In the case of this Zorn, those dark accents in the face float in the middle of a sea of creamy white, the reverse of the usual tonal scheme of a portrait.

Watch a 15-second video clip of my sketch in context on my Instagram, Twitter, or Facebook page.

-----

Related posts: 

Zorn's Two-Source Lighting
Split lighting

Portrait Demo — Three Hours in Three Minutes

If you weren't able catch my portrait demo at the Portrait Society of America conference in Washington, no worries! Here's a front row seat, complete with super-speed action, blow-by-blow narration, and cheat-sheet notes. (Link to YouTube) 

The original is available for silent-auction sale—if you're interested in it, maybe a friend of yours at the convention can put a bid on it for you. (Check Facebook or #artoftheportrait2016 on Twitter or Instagram.)

Portrait Society Conference

Photo by Caleb

I'm at the Portrait Society of America's annual conference in Washington, DC, and was honored yesterday to participate in the Face-Off event, where 15 artists paint three-hour portraits from five models. It was fun to get the oils going again, after painting so much in water media lately. 

Other Face-Off participants included Carol Arnold, Anna Rose Bain, Judith Carducci, Casey Childs, Romel de la Torre, Michelle Dunaway, Max Ginsburg, Quang Ho, Robert Liberace, Ricky Mujica, Teresa Oaxaca, Alicia Ponzio, and Elizabeth Zanzinger.

I shot some time lapse and live video clips, and I'll share those with you on a future post.

Pros and Cons of Posting Work-in-Progress

There are a lot of ways to share your process online, especially after a painting is finished.

But what a lot of people do is show a work developing over time from its early stages to the finish. When artists do that, we can follow stages of their paintings or even their big book projects over a time scale of days, weeks, or months.

Is it a good idea to post unfinished work online? I can think of a few pros and cons, and would love to hear your thoughts.

Pros
1. It's a good way to involve followers in a work that you're doing. It's the essence of Kickstarter campaigns.
2. It's exciting for viewers to watch the ups and downs of the work in real time as it develops.
3. It's valuable for students, fans, and collectors to learn about your process.

Cons
1. You lose the impact of the first impression, something you can never get back. Movie companies never share their storyboards; they only tease with impressive finished clips. Unless there's something incredibly compelling about your process, don't take us behind the curtain until after we've seen the finished thing.

2. A painting goes through some pretty awful stages (at least mine do) and some finishes aren't fit to be shared. No reason to lower my average and clog everyone's feed by dumping all my intermediate stages onto social media.

3. If you put a lot of standalone JPEGs of unfinished works on your blog or Pinterest, they'll come up on a search, and the person searching may assume it's a finish (or just a bad painting). This is less of a problem with media that are more opaque to search, such as Twitter, Instagram, and Snapchat.

Disrupting Face Recognition Technology

Face recognition systems are getting very good at spotting a face in a crowd, whether using cameras in public places or software scanning millions of images posted to social media.

Facebook's DeepFace system, for example, uses a neural-net based machine learning approach to identify any person with 97 percent accuracy.

Faces identified in red. Green square indicates no face detected. Via CVDazzle

Ordinary disguises, such as hats, wigs, mustaches, and glasses may fool human observers, but they don't trick these machine systems. However, face recognition systems only work when they see something that looks like a face.

In a new confluence of couture fashion and privacy activism, a group of hackers is adopting the methods of dazzle camouflage to disrupt face-detection technology.

Contrasting patterns of light and dark cross the contours of the face and overlap the features, making it hard to recognize the shape of the head, and interfering with the edges of the facial features. 

Note to concept artists: these styles would fit well into a futuristic cyberpunk world.

Random patterns are painted onto the face. Hair is alternately curly and straight. 

Whether these fashions are accepted or ultimately banned in public places is anyone's guess, since authorities will argue that terrorists can use them too. 

Such methods may only be effective temporarily, since machine learning systems are now being applied to individual movements, such as gesture and gait recognition, with comparable levels of accuracy.

Yet such advances in technology—and countermeasures to that technology—call into question our basic human assumptions about the expectation of privacy and anonymity in public places.

----
Read more: Camouflage from Face Recognition Technology
Scientific papers on deep learning systems for movement recognition
Previous post: Dazzle Camouflage
Wikipedia on Facebook's DeepFace system

Admissions Officer

I love sketching people when they're moving and talking, rather than posing. I sketched this admissions officer at a meeting for prospective college students. The challenge was to choose a characteristic angle and pose as he spoke.

I'll be lecturing and demoing about sketching people in the wild at the Portrait Society of America's National Conference in Washington, DC this weekend. I don't know if you can still register to attend, but if you can make it, I recommend it highly. Here's the registration form.

During my talk Friday morning called "Likeness and Character," I'll be previewing a segment from an upcoming video that I'm working on called "Portraits in the Wild."

Chris Watson's Audio Postcards

If you like listening to immersive soundscapes while you're painting, you'll enjoy the BBC podcasts by Chris Watson. He's a wildlife sound recordist who takes his sensitive equipment all over the world. His 15-minute programs punctuate the environmental captures with his voice identifying what you're hearing.

Sample episodes:
Midnight at the Oasis—The sounds of the Kalahari Desert, from dusk to dawn, including interesting audio perspectives from the microphone beneath the sand dunes and under the bark of trees.

St James Park—Tracking wildlife in urban environments near his home in Newcastle-on-Tyne, England, he starts with the weekend revelers and sports fans discarding food and litter. Then we hear the sound of rats and mice eating up the scraps, followed by the urban predators, such as tawny owls and foxes.

Glacial Melt — The sounds of calving glaciers in Antarctica, together with the birds, seals, and whales as heard from above and below the water. Watson combines a rich imagination to the informed awareness of a naturalist.

The Ghost Roost — Various sound perspectives tell the story of a giant flock of starlings that arrives to occupy an abandoned pier at West Brighton.
-----
More Chris Watson sound programs
He has contributed audio to the BBC The Life of Birds video documentaries.
Watson talks with David Attenborough about their lives in sound (28 minutes)

Church in the Rain

I just finished writing an article for International Artist magazine which will be out in a few weeks. The subject is painting when you're stuck somewhere. In this case I was in Millbridge, Maine during a camping trip, waiting for the rain to stop. 

I had my painting gear but no umbrella. The only public place with cover was the porch in front of the post office. 

The view looked across to the Congregational church beyond some utility poles and outbuildings. I liked the way the church was white against the white of the sky, with a few birds perched up high on the steeple.

Cardboard Cutout for Reference

James Warhola posing for an Etruscan musician

Here's a tip when you need a quick prop for your model to interact with, but you don't have time to build an elaborate replica. Just draw the shape of it on the surface of a piece of cardboard and cut out what you need. At least that gives the model something three dimensional to hold, and you don't have to make up the whole thing.

It works for a lot of things—instruments, guns, shields, furniture, windows. It only takes a couple of minutes to build, and it's recyclable when you're done.

The full illustration appears in the June, 1988 issue of National Geographic, in the story "The Eternal Etruscans" or on the History and Science section of my website.
It also appears in my book Imaginative Realism: How to Paint What Doesn't Exist