In the past couple of months, UnifyID has been busy attending university hackathons at MIT and UC Berkeley. What this means is hours and hours of non-stop hacking, but it also means unlimited snacks, mini midnight workouts, and lots of young, passionate coders working to create impactful projects.
John poses with a16z representative Nigel at HackMIT.On September 16, John Whaley flew to Cambridge, Massachusetts to attend HackMIT: Hack to the Future where he had the opportunity to meet more than 1500 students from all different universities. Representing a16z, John participated in a fireside chat where he covered a variety of topics including what it’s like to work in a startup, choosing industry versus graduate school, and building a company on machine learning. He discussed the fundamentals of entrepreneurship, team-building, fundraising, and more, as students picked his brain about technical topics and career advice. Later, John was able to speak more in depth during his tech talk about UnifyID and identifying individuals based on gait. Students were deeply interested in the problem UnifyID is trying to solve as well as the impact and intellectual aspect of UnifyID’s approach to the issue.
Aside from his fireside chat and tech talk, John had the opportunity to mentor hackers in their own projects. His favorite part was meeting and interacting with all of the students, seeing their ambition, passion, and genuine interest in the projects that they were working on. He also enjoyed the intense energy in the arena, choosing to stay and mentor hackers until 3am.
After 24 hours of hard work and non stop hacking at MIT, many groups of students presented their projects. Projects covered a wide range of topics from virtual reality games to homework-help mobile applications. Even though John had been to plenty of hackathons in the past, he was still amazed by the caliber and level of innovation that the students were able to reach in their projects. The first place prize ended up going to a group of students who created Pixelator, “a simple product that sharpens blurry images without a lot of code.”
A few weeks later, on October 6, Andres Castaneda crossed the Bay to attend Cal Hacks 4.0 at the UC Berkeley Stadium. With nearly 1500 students listening, he gave a presentation about UnifyID’s Android SDK and API, receiving a positive response from students who believed it was a revolutionary idea. Similar to John at MIT, Andres also had the opportunity to mentor up-and-coming hackers. For 36 hours, he helped students solve technical challenges as they competed for over $100,000 in prizes, including UnifyID’s contribution: a $300 Amazon giftcard and a Rick and Morty card game.
Based on the level of positive impact, innovation, and technical difficulty, the winning hack for UnifyID’s prize was Safescape, a mobile application that analyzes real-time news articles and alerts people in areas of “non-safe” events. It uses UnifyID’s Android SDK to validate individuals on the application. Inspired by the recent natural and terror crises occurring globally, Safescape also provides those in danger with potential escape routes, allows them to alert others around them, and contains a simple way to contact loved ones.
Andres’ favorite part about participating in Cal Hacks was “seeing people build a product from 0 to 1 in 36 hours.” He also found it hilarious that many students brought sleeping bags and threw them on the floor for intermittent opportunities to take naps.
Andres poses with mentees and previous UnifyID interns Aditya and Michael.
UnifyID is a strong supporter of hackathons because they provide great opportunities to connect with university students. Witnessing the high caliber of work accomplished at these events, UnifyID is inspired by young hackers who are truly passionate about making an impact in the world. These students represent a large diversity of talent from all different schools and backgrounds and are able to demonstrate what students are interested in nowadays. Additionally, hackathons allow UnifyID the chance to give back to the community. They are not only learning opportunities for up-and-coming hackers, but they also help UnifyID to understand how to cater to students’ interests and needs. After 2 hackathons in the span of one month, UnifyID is channeling its focus back to the day-to-day for now; however, we cannot wait for the next one!
Vinay Uday Prabhu and John Whaley, UnifyID, San Francisco, CA 94107
Abstract
In this paper, we would like to draw attention towards the vulnerability of the motion sensor-based gait biometric in deep learning-based implicit authentication solutions, when attacked with adversarial perturbations, obtained via the simple fast-gradient sign method. We also showcase the improvement expected by incorporating these synthetically-generated adversarial samples into the training data.
Introduction
In recent times, password entry-based user-authentication methods have increasingly drawn the ire of the security community [1], especially when it comes to its prevalence in the world of mobile telephony. Researchers [1] recently showcased that creating passwords on mobile devices not only takes significantly more time, but it is also more error prone, frustrating, and, worst of all, the created passwords were inherently weaker. One of the promising solutions that has emerged entails implicit authentication [2] of users based on behavioral patterns that are sensed without the active participation of the user. In this domain of implicit authentication, measurement of gait-cycle [3] signatures, mined using the on-phone Inertial Measurement Unit – MicroElectroMechanical Systems (IMU-MEMS) sensors, such as accelerometers and gyroscopes, has emerged as an extremely promising passive biometric [4, 5, 6]. As stated in [7, 5], gait patterns can not only be collected passively, at a distance, and unobtrusively (unlike iris, face, fingerprint, or palm veins), they are also extremely difficult to replicate due to their dynamic nature.
Inspired by the immense success that Deep Learning (DL) has enjoyed in recent times across disparate domains, such as speech recognition, visual object recognition, and object detection [8], researchers in the field of gait-based implicit authentication are increasingly embracing DL-based machine-learning solutions [4, 5, 6, 9], thus replacing the more traditional hand-crafted-feature- engineering-driven shallow machine-learning approaches [10]. Besides circumventing the oft-contentious process of hand-engineering the features, these DL-based approaches are also more robust to noise [8], which bodes well for the implicit-authentication solutions that will be deployed on mainstream commercial hardware. As evinced in [4, 5], these classifiers have already attained extremely high accuracy (∼96%), when trained under the k-class supervised classification framework (where k pertains to the number of individuals). While these impressive numbers give the impression that gait-based deep implicit authentication is ripe for immediate commercial implementation, we would like to draw the attention of the community towards a crucial shortcoming. In 2014, Szegedy et al. [11] discovered that, quite like shallow machine-learning models, the state-of- the-art deep neural networks were vulnerable to adversarial examples that can be synthetically generated by strategically introducing small perturbations that make the resultant adversarial input example only slightly different from correctly classified examples drawn from the data distribution, but at the same time resulting in a potentially controlled misclassification. To make things worse, a large plethora of models with disparate architectures, trained on different subsets of the training data, have been found to misclassify the same adversarial example, uncovering the presence of fundamental blind spots in our DL frameworks. After this discovery, several works have emerged ([12, 13]), addressing both means of defence against adversarial examples, as well as novel attacks. Recently, the cleverhans software library [13] was released. It provides standardized reference implementations of adversarial example-construction techniques and adversarial training, thereby facilitating rapid development of machine-learning models, robust to adversarial attacks, as well as providing standardized benchmarks of model performance in the adversarial setting explained above. In this paper, we focus on harnessing the simplest of all adversarial attack methods, i.e. the fast gradient sign method (FGSM) to attack the IDNet deep convolutional neural network (DCNN)-based gait classifier introduced in [4]. Our main contributions are as follows: 1: This is, to the best of our knowledge, the first paper that introduces deep adversarial attacks into this non-computer vision setting, specifically, the gait-driven implicit-authentication domain. In doing so, we hope to draw the attention of the community towards this crucial issue in the hope that further publications will incorporate adversarial training as a default part of their training pipelines. 2: One of the enduring images that is widely circulated in adversarial training literature is that of the panda+nematode = gibbon adversarial-attack example on GoogleNet in [14], which was instrumental in vividly showcasing the potency of the blind spot. In this paper, we do the same with accelerometric data to illustrate how a small and seemingly imperceptible perturbation to the original signal can cause the DCNN to make a completely wrong inference with high probability. 3: We empirically characterize the degradation of classification accuracy, when subjected to an FGSM attack, and also highlight the improvement in the same, upon introducing adversarial training. 4: Lastly, we have open-sourced the code here.
Figure 1. Variation in the probability of correct classification (37 classes) with and without adversarial training for varying ε.Figure 2. The true accelerometer amplitude signal and its adversarial counterpart for ε = 0.4.
2. Methodology and Results
In this paper, we focus on the DCNN-based IDNet [4] framework, which entails harnessing low-pass-filtered tri-axial accelerometer and gyroscope readings (plus the sensor-specific magnitude signals), to, firstly, extract the gait template, of dimension 8 × 200, which is then used to train a DCNN in a supervised-classification setting. In the original paper, the model identified users in real time by using the DCNN as a deep-feature extractor and further training an outlier detector (one-class support vector machine-SVM), whose individual gait-wise outputs were finally combined into a Wald’s probability-ratio-test-based framework. Here, we focus on the trained IDNet-DCNN and characterize its performance in the adversarial-training regime. To this end, we harness the FGSM introduced in [14], where the adversarial example, x ̃, for a given input sample, x, is generated by: x ̃ = x + ε sign (∇xJ (θ, x)), where θ represents the parameter vector of the DCNN, J (θ, x) is the cost function used to train the DCNN, and ∇x () is the gradient function.
As seen, this method is parametrized by ε, which controls the magnitude of the inflicted perturbations. Fig. 2 showcases the true and adversarial gait-cycle signals for the accelerometer magnitude signal (given by amag(t) = √(a2x (t) + a2y (t) + a2z (t))) for ε = 0.4. Fig. 1 captures the drop in the probability of correct classification (37 classes) with increasing ε. First, we see that in the absence of any adversarial example, we were able to get about 96% ac- curacy on a 37 class classification problem, which is very close to what is claimed in [4]. However, with even mild perturbations (ε = 0.4), we see a sharp decrease of nearly 40% in accuracy. Fig. 1 also captures the effect of including the synthetically generated adversarial examples in this scenario. We see that, for ε = 0.4, we manage to achieve about 82% accuracy, which is a vast improvement of ∼ 25%.
3. Future Work
This brief paper is part of an ongoing research endeavor. We are currently currently extending this work to other adversarial-attack approaches, such as Jacobian-based Saliency-Map Approach (JSMA) and Black-Box-Attack (BBA) approach [15]. We are also investigating the effect of these attacks within the deep-feature-extraction+SVM approach of [4], and we are comparing other architectures, such as [6] and [5].
References
[1] W.Melicher, D.Kurilova, S.M.Segreti, P.Kalvani, R.Shay, B. Ur, L. Bauer, N. Christin, L. F. Cranor, and M. L. Mazurek, “Usability and security of text passwords on mobile devices,” in Proceedings of the 2016 CHI Conference on Human Factors in Computing Systems, pp. 527–539, ACM, 2016. 1
[2] E. Shi, Y. Niu, M. Jakobsson, and R. Chow, “Implicit authentication through learning user behavior,” in International Conference on Information Security, pp. 99–113, Springer, 2010. 1
[3] J. Perry, J. R. Davids, et al., “Gait analysis: normal and pathological function.,” Journal of Pediatric Orthopaedics, vol. 12, no. 6, p. 815, 1992. 1
[4] M. Gadaleta and M. Rossi, “Idnet: Smartphone-based gait recognition with convolutional neural networks,” arXiv preprint arXiv:1606.03238, 2016. 1, 2
[5] Y. Zhao and S. Zhou, “Wearable device-based gait recognition using angle embedded gait dynamic images and a convolutional neural network,” Sensors, vol. 17, no. 3, p. 478, 2017. 1, 2
[6] S. Yao, S. Hu, Y. Zhao, A. Zhang, and T. Abdelza- her, “Deepsense: A unified deep learning framework for time-series mobile sensing data processing,” arXiv preprint arXiv:1611.01942, 2016. 1, 2
[7] S. Wang and J. Liu, Biometrics on mobile phone. INTECH Open Access Publisher, 2011. 1
[8] Y. LeCun, Y. Bengio, and G. Hinton, “Deep learning,” Nature, vol. 521, no. 7553, pp. 436–444, 2015. 1
[9] N. Neverova, C. Wolf, G. Lacey, L. Fridman, D. Chandra, B. Barbello, and G. Taylor, “Learning human identity from motion patterns,” IEEE Access, vol. 4, pp. 1810–1820, 2016. 1
[10] C. Nickel, C. Busch, S. Rangarajan, and M. Mo ̈bius, “Using hidden markov models for accelerometer-based biometric gait recognition,” in Signal Processing and its Applications (CSPA), 2011 IEEE 7th International Colloquium on, pp. 58–63, IEEE, 2011. 1
[11] C. Szegedy, W. Zaremba, I. Sutskever, J. Bruna, D. Erhan, I. Goodfellow, and R. Fergus, “Intriguing properties of neural networks,” arXiv preprint arXiv:1312.6199, 2013. 1
[12] C. Szegedy, W. Liu, Y. Jia, P. Sermanet, S. Reed, D. Anguelov, D. Erhan, V. Vanhoucke, and A. Rabinovich, “Going deeper with convolutions,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 1–9, 2015. 1
[13] N. Papernot, I. Goodfellow, R. Sheatsley, R. Feinman, and P. McDaniel, “cleverhans v1.0.0: an adversarial machine learning library,” arXiv preprint arXiv:1610.00768, 2016. 1
[14] I. J. Goodfellow, J. Shlens, and C. Szegedy, “Explain- ing and harnessing adversarial examples,” arXiv preprint arXiv:1412.6572, 2014. 2
[15] N. Papernot, P. McDaniel, I. Goodfellow, S. Jha, Z. B. Celik, and A. Swami, “Practical black-box attacks against deep learning systems using adversarial examples,” arXiv preprint arXiv:1602.02697, 2016.
Vinay Uday Prabhu and John Whaley, UnifyID, San Francisco, CA 94107
Abstract
In this paper, we demonstrate a simple face spoof attack targeting the face recognition system of a widely available commercial smart-phone. The goal of this paper is not proclaim a new spoof attack but to rather draw the attention of the anti-spoofing researchers towards a very specific shortcoming shared by one-shot face recognition systems that involves enhanced vulnerability when a smiling reference image is used.
Introduction
One-shot face recognition (OSFR) or single sample per person (SSPP) face recognition is a well-studied research topic in computer vision (CV) [8]. Solutions such as Local Binary Pattern (LBP) based detectors [1], Deep Lambertian Networks (DLN) [9] and Deep Supervised Autoencoders (DSA) [4] have been proposed in recent times to make the OSFR system more robust to changes in illumination, pose, facial expression and occlusion that they encounter when deployed in the wild. One very interesting application of face recognition that has gathered traction lately is for mobile device unlocking [6]. One of the highlights of Android 4.0 (Ice Cream Sandwich) was the Face Unlock screen-lock option that allowed users to unlock their devices with their faces. It is rather imperative that we mention here that this option is always presented to the user with a cautioning clause that typically reads like *Face recognition is less secure than pattern, PIN, or password.
The reasoning behind this is that there exists a plethora of face spoof attacks such as print attacks, malicious identical twin attack, sleeping user attack, replay attacks and 3D mask attacks. These attacks are all fairly successful against most of the commercial off-the-shelf face recognizers [7]. This ease of spoof attacks has also attracted attention of the CV researchers that has led to a lot of efforts in developing liveness detection anti-spoofing frameworks such as Secure-face [6]. (See [3] for a survey.)
Recently, a large scale smart-phone manufacturer introduced a face recognition based phone unlocking feature. This announcement was promptly followed by media reports about users demonstrating several types of spoof attacks.
In this paper, we would like to explore a simple print attack on this smart-phone. The goal of this paper is not proclaim a new spoof attack but to rather draw the attention of the anti-spoofing community towards a very specific shortcoming shared by face recognition systems that we uncovered in this investigation.
2. Methodology and Results
Figure 1. Example of two neutral expression faces that failed to spoof the smart-phone’s face recognition system.Figure 2. Example of 2 smiling registering faces that successfully spoofed the smart-phone’s face recognition system.
The methodology we used entailed taking a low quality printout of the target user’s face on a plain white US letter paper size (of dimension 8.5 by 11.0 inches) and then unlocking the device by simply exposing this printed paper in front of the camera. Given the poor quality of the printed images, we observed that this simple print attack was duly repulsed by the detector system as long as the attacker sported neutral facial expressions during the registration phase. However, when we repeated the attack in such a way that the attacker had an overtly smiling face when (s)he registered, we were able to break in successfully with high regularity.
In Figure 1, we see two examples of neutral expression faces that failed to spoof the smart-phone’s face recognition system when the registering image had a neutral facial expression.
In Figure 2, we see the same two subjects’ images that successfully spoofed the phone’s face recognition system when the registering (enrollment) image was overtly smiling.
2.1. Motivation for the attack and discussion
It has been well known for a long time in the computer vision community that faces displaying expressions, especially smiles, resulted in stronger recall and discrimination power [10]. In fact, the authors in [2] termed this the happy-face advantage, and showcased the variation in detection performance for varying facial expressions. Through experimentation, we wanted to investigate the specific onshot classification scenario when the registering enrollment face had a strong smile that resulted in the discovery of this attack. As for defense from this attack, there are two straightforward recommendations. The first recommendation would be to simply display a message goading the user to maintain a passport-type neutral facial expression. The second would entail having a smile detector such as [5] as a pre-filter that would only allow smile-free images as a reference image.
References
[1] T. Ahonen, A. Hadid, and M. Pietikainen. Face description with local binary patterns: Application to face recognition. IEEE transactions on pattern analysis and machine intelligence, 28(12):2037–2041, 2006. 1
[2] W. Chen, K. Lander, and C. H. Liu. Matching faces with emotional expressions. Frontiers in psychology, 2:206, 2011. 2
[3] J. Galbally, S. Marcel, and J. Fierrez. Biometric antispoofing methods: A survey in face recognition. IEEE Access, 2:1530–1552, 2014. 1
[4] S. Gao, Y. Zhang, K. Jia, J. Lu, and Y. Zhang. Single sample face recognition via learning deep supervised autoencoders. IEEE Transactions on Information Forensics and Security, 10(10):2108–2118, 2015. 1
[5] P. O. Glauner. Deep convolutional neural networks for smile recognition. arXiv preprint arXiv:1508.06535, 2015. 2
[6] K. Patel, H. Han, and A. K. Jain. Secure face unlock: Spoof detection on smartphones. IEEE Transactions on Information Forensics and Security, 11(10):2268–2283, 2016. 1
[7] D. F. Smith, A. Wiliem, and B. C. Lovell. Face recognition on consumer devices: Reflections on replay attacks. IEEE Transactions on Information Forensics and Security,10(4):736–745, 2015. 1
[8] X.Tan,S.Chen,Z.-H.Zhou, and F.Zhang. Face recognition from a single image per person: A survey. Pattern recognition, 39(9):1725–1745, 2006.
[9] Y. Tang, R. Salakhutdinov, and G. Hinton. Deep lambertian networks. arXiv preprint arXiv:1206.6445, 2012. 1
[10] Y. Yacoob and L. Davis. Smiling faces are better for face recognition. In Automatic Face and Gesture Recognition, 2002. Proceedings. Fifth IEEE International Conference on, pages 59–64. IEEE, 2002. 2
