arXiv subjects

Yingying Chen

Publications and source records attributed to Yingying Chen.

48 records · Page 3Linked to original sources

arXiv2021

Enabling Fast and Universal Audio Adversarial Attack Using Generative Model

Recently, the vulnerability of DNN-based audio systems to adversarial attacks has obtained the increasing attention. However, the existing audio adversarial attacks allow the adversary to possess the entire user's audio input as well as granting sufficient time budget to generate the adversarial perturbations. These idealized assumptions, however, makes the existing audio adversarial attacks mostly impossible to be launched in a timely fashion in practice (e.g., playing unnoticeable adversarial perturbations along with user's streaming input). To overcome these limitations, in this paper we propose fast audio adversarial perturbation generator (FAPG), which uses generative model to generate adversarial perturbations for the audio input in a single forward pass, thereby drastically improving the perturbation generation speed. Built on the top of FAPG, we further propose universal audio adversarial perturbation generator (UAPG), a scheme crafting universal adversarial perturbation that can be imposed on arbitrary benign audio input to cause misclassification. Extensive experiments show that our proposed FAPG can achieve up to 167X speedup over the state-of-the-art audio adversarial attack methods. Also our proposed UAPG can generate universal adversarial perturbation that achieves much better attack performance than the state-of-the-art solutions.

eess.AS↗

arXiv2020

Motion Sensor-based Privacy Attack on Smartphones

In this paper, we build a speech privacy attack that exploits speech reverberations generated from a smartphone's in-built loudspeaker captured via a zero-permission motion sensor (accelerometer). We design our attack Spearphone2, and demonstrate that speech reverberations from inbuilt loudspeakers, at an appropriate loudness, can impact the accelerometer, leaking sensitive information about the speech. In particular, we show that by exploiting the affected accelerometer readings and carefully selecting feature sets along with off-the-shelf machine learning techniques, Spearphone can successfully perform gender classification (accuracy over 90%) and speaker identification (accuracy over 80%) for any audio/video playback on the smartphone. Our results with testing the attack on a voice call and voice assistant response were also encouraging, showcasing the impact of the proposed attack. In addition, we perform speech recognition and speech reconstruction to extract more information about the eavesdropped speech to an extent. Our work brings to light a fundamental design vulnerability in many currently-deployed smartphones, which may put people's speech privacy at risk while using the smartphone in the loudspeaker mode during phone calls, media playback or voice assistant interactions.

cs.CR↗

arXiv2020

Learning Feature Embeddings for Discriminant Model based Tracking

After observing that the features used in most online discriminatively trained trackers are not optimal, in this paper, we propose a novel and effective architecture to learn optimal feature embeddings for online discriminative tracking. Our method, called DCFST, integrates the solver of a discriminant model that is differentiable and has a closed-form solution into convolutional neural networks. Then, the resulting network can be trained in an end-to-end way, obtaining optimal feature embeddings for the discriminant model-based tracker. As an instance, we apply the popular ridge regression model in this work to demonstrate the power of DCFST. Extensive experiments on six public benchmarks, OTB2015, NFS, GOT10k, TrackingNet, VOT2018, and VOT2019, show that our approach is efficient and generalizes well to class-agnostic target objects in online tracking, thus achieves state-of-the-art accuracy, while running beyond the real-time speed. Code will be made available.

cs.CV↗

arXiv2020

Real-time, Universal, and Robust Adversarial Attacks Against Speaker Recognition Systems

As the popularity of voice user interface (VUI) exploded in recent years, speaker recognition system has emerged as an important medium of identifying a speaker in many security-required applications and services. In this paper, we propose the first real-time, universal, and robust adversarial attack against the state-of-the-art deep neural network (DNN) based speaker recognition system. Through adding an audio-agnostic universal perturbation on arbitrary enrolled speaker's voice input, the DNN-based speaker recognition system would identify the speaker as any target (i.e., adversary-desired) speaker label. In addition, we improve the robustness of our attack by modeling the sound distortions caused by the physical over-the-air propagation through estimating room impulse response (RIR). Experiment using a public dataset of 109 English speakers demonstrates the effectiveness and robustness of our proposed attack with a high attack success rate of over 90%. The attack launching time also achieves a 100X speedup over contemporary non-universal attacks.

eess.AS↗

arXiv2020

EchoLock: Towards Low Effort Mobile User Identification

User identification plays a pivotal role in how we interact with our mobile devices. Many existing authentication approaches require active input from the user or specialized sensing hardware, and studies on mobile device usage show significant interest in less inconvenient procedures. In this paper, we propose EchoLock, a low effort identification scheme that validates the user by sensing hand geometry via commodity microphones and speakers. These acoustic signals produce distinct structure-borne sound reflections when contacting the user's hand, which can be used to differentiate between different people based on how they hold their mobile devices. We process these reflections to derive unique acoustic features in both the time and frequency domain, which can effectively represent physiological and behavioral traits, such as hand contours, finger sizes, holding strength, and gesture. Furthermore, learning-based algorithms are developed to robustly identify the user under various environments and conditions. We conduct extensive experiments with 20 participants using different hardware setups in key use case scenarios and study various attack models to demonstrate the performance of our proposed system. Our results show that EchoLock is capable of verifying users with over 90% accuracy, without requiring any active input from the user.

cs.HC↗

arXiv2020

WearID: Wearable-Assisted Low-Effort Authentication to Voice Assistants using Cross-Domain Speech Similarity

Due to the open nature of voice input, voice assistant (VA) systems (e.g., Google Home and Amazon Alexa) are under a high risk of sensitive information leakage (e.g., personal schedules and shopping accounts). Though the existing VA systems may employ voice features to identify users, they are still vulnerable to various acoustic attacks (e.g., impersonation, replay and hidden command attacks). In this work, we focus on the security issues of the emerging VA systems and aim to protect the users' highly sensitive information from these attacks. Towards this end, we propose a system, WearID, which uses an off-the-shelf wearable device (e.g., a smartwatch or bracelet) as a secure token to verify the user's voice commands to the VA system. In particular, WearID exploits the readily available motion sensors from most wearables to describe the command sound in vibration domain and check the received command sound across two domains (i.e., wearable's motion sensor vs. VA device's microphone) to ensure the sound is from the legitimate user.

cs.HC↗

arXiv2020

WiEat: Fine-grained Device-free Eating Monitoring Leveraging Wi-Fi Signals

Eating is a fundamental activity in people's daily life. Studies have shown that many health-related problems such as obesity, diabetes and anemia are closely associated with people's unhealthy eating habits (e.g., skipping meals, eating irregularly and overeating). Traditional eating monitoring solutions relying on self-reports remain an onerous task, while the recent trend requiring users to wear expensive dedicated hardware is still invasive. To overcome these limitations, in this paper, we develop a device-free eating monitoring system using WiFi-enabled devices (e.g., smartphone or laptop). Our system aims to automatically monitor users' eating activities through identifying the fine-grained eating motions and detecting the chewing and swallowing. In particular, our system extracts the fine-grained Channel State Information (CSI) from WiFi signals to distinguish eating from non-eating activities and further recognizing users' detailed eating motions with different utensils (e.g., using a folk, knife, spoon or bare hands). Moreover, the system has the capability of identifying chewing and swallowing through detecting users' minute facial muscle movements based on the derived CSI spectrogram. Such fine-grained eating monitoring results are beneficial to the understanding of the user's eating behaviors and can be used to estimate food intake types and amounts. Extensive experiments with 20 users over 1600-minute eating show that the proposed system can recognize the user's eating motions with up to 95% accuracy and estimate the chewing and swallowing amount with 10% percentage error.

cs.HC↗

arXiv2020

BDD100K: A Diverse Driving Dataset for Heterogeneous Multitask Learning

Datasets drive vision progress, yet existing driving datasets are impoverished in terms of visual content and supported tasks to study multitask learning for autonomous driving. Researchers are usually constrained to study a small set of problems on one dataset, while real-world computer vision applications require performing tasks of various complexities. We construct BDD100K, the largest driving video dataset with 100K videos and 10 tasks to evaluate the exciting progress of image recognition algorithms on autonomous driving. The dataset possesses geographic, environmental, and weather diversity, which is useful for training models that are less likely to be surprised by new conditions. Based on this diverse dataset, we build a benchmark for heterogeneous multitask learning and study how to solve the tasks together. Our experiments show that special training strategies are needed for existing models to perform such heterogeneous tasks. BDD100K opens the door for future studies in this important venue.

cs.CV↗

arXiv2018

Recurrent Calibration Network for Irregular Text Recognition

Scene text recognition has received increased attention in the research community. Text in the wild often possesses irregular arrangements, typically including perspective text, curved text, oriented text. Most existing methods are hard to work well for irregular text, especially for severely distorted text. In this paper, we propose a novel Recurrent Calibration Network (RCN) for irregular scene text recognition. The RCN progressively calibrates the irregular text to boost the recognition performance. By decomposing the calibration process into multiple steps, the irregular text can be calibrated to normal one step by step. Besides, in order to avoid the accumulation of lost information caused by inaccurate transformation, we further design a fiducial-point refinement structure to keep the integrity of text during the recurrent process. Instead of the calibrated images, the coordinates of fiducial points are tracked and refined, which implicitly models the transformation information. Based on the refined fiducial points, we estimate the transformation parameters and sample from the original image at each step. In this way, the original character information is preserved until the final transformation. Such designs lead to optimal calibration results to boost the performance of succeeding recognition. Extensive experiments on challenging datasets demonstrate the superiority of our method, especially on irregular benchmarks.

cs.CV↗

arXiv2017

Reading Scene Text with Attention Convolutional Sequence Modeling

Reading text in the wild is a challenging task in the field of computer vision. Existing approaches mainly adopted Connectionist Temporal Classification (CTC) or Attention models based on Recurrent Neural Network (RNN), which is computationally expensive and hard to train. In this paper, we present an end-to-end Attention Convolutional Network for scene text recognition. Firstly, instead of RNN, we adopt the stacked convolutional layers to effectively capture the contextual dependencies of the input sequence, which is characterized by lower computational complexity and easier parallel computation. Compared to the chain structure of recurrent networks, the Convolutional Neural Network (CNN) provides a natural way to capture long-term dependencies between elements, which is 9 times faster than Bidirectional Long Short-Term Memory (BLSTM). Furthermore, in order to enhance the representation of foreground text and suppress the background noise, we incorporate the residual attention modules into a small densely connected network to improve the discriminability of CNN features. We validate the performance of our approach on the standard benchmarks, including the Street View Text, IIIT5K and ICDAR datasets. As a result, state-of-the-art or highly-competitive performance and efficiency show the superiority of the proposed approach.

cs.CV↗

arXiv2017

Fast Deep Matting for Portrait Animation on Mobile Phone

Image matting plays an important role in image and video editing. However, the formulation of image matting is inherently ill-posed. Traditional methods usually employ interaction to deal with the image matting problem with trimaps and strokes, and cannot run on the mobile phone in real-time. In this paper, we propose a real-time automatic deep matting approach for mobile devices. By leveraging the densely connected blocks and the dilated convolution, a light full convolutional network is designed to predict a coarse binary mask for portrait images. And a feathering block, which is edge-preserving and matting adaptive, is further developed to learn the guided filter and transform the binary mask into alpha matte. Finally, an automatic portrait animation system based on fast deep matting is built on mobile devices, which does not need any interaction and can realize real-time matting with 15 fps. The experiments show that the proposed approach achieves comparable results with the state-of-the-art matting solvers.

cs.CV↗

arXiv2017

Joint Background Reconstruction and Foreground Segmentation via A Two-stage Convolutional Neural Network

Foreground segmentation in video sequences is a classic topic in computer vision. Due to the lack of semantic and prior knowledge, it is difficult for existing methods to deal with sophisticated scenes well. Therefore, in this paper, we propose an end-to-end two-stage deep convolutional neural network (CNN) framework for foreground segmentation in video sequences. In the first stage, a convolutional encoder-decoder sub-network is employed to reconstruct the background images and encode rich prior knowledge of background scenes. In the second stage, the reconstructed background and current frame are input into a multi-channel fully-convolutional sub-network (MCFCN) for accurate foreground segmentation. In the two-stage CNN, the reconstruction loss and segmentation loss are jointly optimized. The background images and foreground objects are output simultaneously in an end-to-end way. Moreover, by incorporating the prior semantic knowledge of foreground and background in the pre-training process, our method could restrain the background noise and keep the integrity of foreground objects at the same time. Experiments on CDNet 2014 show that our method outperforms the state-of-the-art by 4.9%.

cs.CV↗