CV Dazzle is a type of camouflage from computer vision. It uses bold patterning to break apart the expected features targeted by computer vision algorithms and proves that faces, or other objects, can exist in a dual perceptual state: visible to humans yet invisible to machines. CV Dazzle was developed as my thesis at New York University’s Interactive Telecommunications Program and first published in April 2010.

This page is an overview of the project concepts, motivation, and initial tests. More technical information about reverse engineering and visualizing the vulnerabilities in the haarcascade and other computer vision algorithms may be published here in the future.

How It Works #

The initial CV Dazzle designs work by altering the expected dark and light areas of a face (or object) according to the vulnerabilities of a specific computer vision algorithm. In the image above (Look #5), the design targets the Viola-Jones face detection algorithm, a popular (at the time of development) and open source face detector that is included with the OpenCV computer vision framework. But CV Dazzle is not a specific design or pattern. CV Dazzle is a camouflage strategy to evade computational vision systems that evolves, as camouflage does, alongside the technology it aims to subvert. Patterns and designs are always specific to the wearer, algorithm, and environmental conditions.

Designs can be created using only hair styling, makeup, and fashion accessories, which could be customized to any wearer’s style and are low-cost or free, and accessible to a wide audience. Newer forms of a CV Dazzle approach could target other algorithms, such as deep convolutional neural networks, but would require finding vulnerabilities in these algorithms. Because computer vision is a probabilistic determination, finding the right look is about finding how to appear one step below the threshold of detection.

Since face detection is the first step in any automated facial recognition system, blocking the detection stage also blocks any subsequent facial analysis including recognition and emotional analysis. Therefore, CV Dazzle could be used to block facial recognition by blocking face detection using only hair styling and makeup. This is the strategy used for the looks on this page, which all target the Viola-Jones haarcascade face detection algorithm. If a face is not detected by the algorithm, it effectively blocks the subsequent recognition algorithms. Often these two algorithms are conflated, but they are entirely different, each with its own set of vulnerabilities.

Many community and activist uses of CV Dazzle have targeted face recognition, instead of detection. This is also a valid approach. Face recognition, like detection, merely computes a probability for how much one face looks like others, typically using a cosine similarity metric with a 512, 1024, or 2048 length feature vector describing the face. Adding facial coverings, makeup, or prosthetic will affect this score to varying degrees. For example,

Look Book #

The first design (Look #1) was created in 2010 as a proof of concept for my masters thesis at ITP NYU. The subsequent 5 looks were created in collaboration with DIS Magazine (Looks 2-4) and for a commission for the New York Times (Looks 5-6). The goal of of these test looks was to investigate the potential for a style that was functional but still within the margins of conceptual fashion.

Evaluation #

To verify the results below, the images can be tested against each of the OpenCV haarcascade profiles to demonstrate their effectiveness. Use the code below to verify the results in the table.

import cv2 as cv
im = cv.imread(filepath)
im_gray = cv.cvtColor(im, cv.COLOR_BGR2GRAY)
cascade = cv.data + "haarcascade_frontalface_default.xml"
classifier = cv.CascadeClassifier(cascade)

faces = classifier.detectMultiScale(
    im_gray,
    scaleFactor=1.05,
    minNeighbors=3,
    minSize=(50, 50)
)

print("Found {} face candidates".format(len(faces))

# Draw rectangles over faces
for (x, y, w, h) in faces:
    cv.rectangle(im, (x, y), (x+w, y+h), (0, 255, 0), 3)

cv.imshow("Result" ,im)
cv.waitKey(0)
cv.destroyAllWindows()

Results for haarcascade_frontalface_default.xml, haarcascade_frontalface_alt.xml, haarcascade_frontalface_alt2.xml, and haarcascade_frontalface_profileface.xml:

Another way to visualize the effectiveness is to use a saliency (heat) map to show which areas of the facial region are missed or activated by the face detection algorithm. The saliency map is generated by separating each stage of the haarcascade classifier and measuring the confidence score computed for that stage, then merging the sliding window regions to produce a heat map.

The results illustrate and prove that hair and makeup alone can be used to lower the wearer’s probability below the threshold of detection for most of the OpenCV Haarcascade detection profiles. There are important limitations to keep in mind though.

First, these looks were designed to work against the Viola-Jones Haarcascade face detector in 2D still-images in the visible light spectrum with the pretrained Haarcascade detection profiles. Other face detection algorithms including Lineary Binary Pattern (LBP), Histogram of Oriented Gradients (HOG), Covolutional Neural Networks (CNN), multi-camera 3D-based systems, and multi-spectral imaging systems would require a different strategy. These looks are solely based on the Viola-Jones Haarcacade classifiers.

Second, lighting conditions will cause the results to vary. The pose and illumination in these photos is similar to a biometric enrollment (passport style) photo. Overhead or more direct lighting will change the intensity and location of shadows which will change the detection outcome.

Style Tips #

For the best performance a CV Dazzle look is highly specific to the situation, unique to the wearer, designed for specific algorithm and never replicated. The tips below apply only to the Viola-Jones haarcascade method for face detection.

1. Makeup Avoid enhancers. They amplify key facial features. This makes your face easier to detect. Instead apply makeup that contrasts with your skin tone in unusual tones and directions: light colors on dark skin, dark colors on light skin.
2. Nose Bridge Partially obscure the nose-bridge area. The region where the nose, eyes, and forehead intersect is a key facial feature. This is especially effective against OpenCV’s face detection algorithm.
3. Eyes Partially obscure one or both of the ocular regions. The symmetrical position and darkness of eyes is a key facial feature.
4. Masks Avoid wearing masks as they are illegal in some cities. Instead of concealing your face, modify the contrast, tonal gradients, and spatial relationship of dark and light areas using hair, makeup, and/or unique fashion accessories.
5. Head Research from Ranran Feng and Balakrishnan Prabhakaran at University of Texas, shows that obscuring the elliptical shape of a head can also improve your ability to block face detection. Link: Facilitating fashion camouflage art. Use hair, turtlenecks, or fashion accessories to alter the expected elliptical shape.
6. Asymmetry Face detection algorithms expect symmetry between the left and right sides of the face. By developing an asymmetrical look, you can decrease your probability of being detected.

Commissions #

In December 2013 the New York Times commissioned a new look (Look #5) for an Op-Art feature about facial recognition, introducing the concept to a large audience and proving that face detection and face recognition are not as powerful as agencies claim.

Background #

This project began in 2010 as my masters thesis at NYU ITP as a challenge to the growing power asymmetries in computer vision and, in particular, widespread deployment of facial recognition technology.

The name of the project was inspired by WWI ship camouflage called Dazzle that used cubist-inspired designs to break apart the visual continuity of a battleship in order to conceal its orientation and size. Similarly, CV Dazzle, short for Computer Vision Dazzle, uses bold, graphic designs that break apart the visual continuity of a face. While the end result is still visible to human observers, CV Dazzle degrades the visual comprehension of computer vision systems.

This ongoing project is motivated by a need to reclaim privacy in a world of increased visual surveillance and data collection. Computer vision poses new challenges that otherwise did not exist in human observation; it is scalable, remote, networked, magnified, and multi-spectral.

Ideally, there would be a way to appear visible to human observers but less visible to computer vision surveillance systems. This is the goal of CV Dazzle; to mitigate the risks of remote and computational visual information capture and analysis under the guise of fashion. Since beginning this project in 2010, the concerns of a widespread facial recognition have only become more urgent and apparent and hopefully this project will continue to develop.

Responses #

There has been a surprisingly positive response to the CV Dazzle project since first publishing it 2010. Activist and art groups have done workshops to explore and expose the vulnerabilities of biometric recognition systems. Both adults and children have done makeup parties. Fashion designers, musicians, political activists, and hackers have modified or borrowed concepts and taken the original ideas into weird new places. It’s encouraging to see how concepts spread over distance and time, and to see a steadily growing resistance to face recognition.

Fact Check #

There have been some misrepresentations of the project as well. Without drawing further attention to any of the specific writings, I’d like to address several of the most common misrepresentations that I’ve noticed.

The first is that CV Dazzle doesn’t work. Articles making this claim often either try applying a lot of makeup and hairstyle patterning based on the original design, or write about others who did, and expect it to break current face recognition systems. This is a misunderstanding of the project, an oversimplification of the CV Dazzle concept, and a media spectacle designed to fail as a form entertainment. Wearing a lot of makeup is probably not going to break any algorithm unless it is designed to exploit specific vulnerabilities in a specific algorithm . In order to make a claim whether or not a design works, the computer vision algorithm needs to be specified. CV Dazzle is not a specific design; it’s a concept. The concept was proven in 2010 and can theoretically be extended to any computer vision algorithm. For example, here’s a video of the former director of the United States Intelligence Advanced Research Projects Activity (IARPA) in 2017 (7 years after CV Dazzle was introduced) describing his interest in “fairly easy ways to spoof facial recognition” by “you know, taking a magic maker and putting a couple dots on your forehead.”

The second misconception is that the original CV Dazzle designs were supposed to break face recognition. As noted above, the original CV Dazzle designs targeted “face detection” algorithms, not “face recognition”. Face detection and face recognition are completely different algorithms, though they are often conflated by non-technical writers. It is possible to design a look against face detection or face recognition, but also emotional recognition, age estimation, cardiopulmonary analysis, iris recognition, gait recognition, body pose estimation, and any other biometric computer vision algorithm. But that algorithm should be clearly specified to a reader to prevent further misunderstandings. Any article failing to mention a specific algorithm should be dismissed.

The third misconception is that CV Dazzle is outdated. This incorrectly assumes that CV Dazzle is a specific design or look. As noted, CV Dazzle is a concept not a pattern or a product. The concept is that bold appearances can break computer vision algorithms while remaining visible to human perceptual systems. Designs are always specific to algorithms. As technology evolves, designs will become outdated, as do many forms of camouflage (and fashion). At best, camouflage and specific CV Dazzle designs offer a temporary advantage. Also, once the design is revealed it becomes less effective. This is why new designs are no longer published on cvdazzle.com. Sharing a look burns its value. The original designs were presented as proof that the concept is functional, not merely speculative.

Some critics have pointed out that wearing bold makeup in broad daylight is not a practical way to evade surveillance as it would draw more attention. In broad daylight perhaps, but at a party it would probably blend in well. In fact CV Dazzle was originally envisioned for club culture and partially inspired by photos from the Boombox party scene in London. Fashion, like counter-surveillance, evolves over time. What is foreign today could be normal tomorrow, and passé the following day. Fashion is a reflection of our collective desires and fears. But it’s likely that counter-surveillance fashion and new forms of technical camouflage are here to stay.

Another occasional criticism is that CV Dazzle is too individualistic of an approach to counter-surveillance. This attitude should be avoided because it discourages experimenting and probing opaque algorithmic systems. It’s important to question these systems, attempt to break them, and explore them on your own terms. More effort is needed to counter mass surveillance from both individuals and groups. Supporting one should not negate other. Additionally, collective action can emerge from shared personal affinities.

And the last is that counter-surveillance projects somehow normalize surveillance. This is defeatist. Does protesting normalize oppression?