Nilay Sanghvi

Bio: Nilay Sanghvi is an academic researcher from Indraprastha Institute of Information Technology. The author has contributed to research in topics: Deep learning & Vulnerability (computing). The author has an hindex of 1, co-authored 2 publications receiving 2 citations.

MixNet for Generalized Face Presentation Attack Detection

Abstract: The non-intrusive nature and high accuracy of face recognition algorithms have led to their successful deployment across multiple applications ranging from border access to mobile unlocking and digital payments. However, their vulnerability against sophisticated and cost-effective presentation attack mediums raises essential questions regarding its reliability. In the literature, several presentation attack detection algorithms are presented; however, they are still far behind from reality. The major problem with existing work is the generalizability against multiple attacks both in the seen and unseen setting. The algorithms which are useful for one kind of attack (such as print) perform unsatisfactorily for another type of attack (such as silicone masks). In this research, we have proposed a deep learning-based network termed as \textit{MixNet} to detect presentation attacks in cross-database and unseen attack settings. The proposed algorithm utilizes state-of-the-art convolutional neural network architectures and learns the feature mapping for each attack category. Experiments are performed using multiple challenging face presentation attack databases such as SMAD and Spoof In the Wild (SiW-M) databases. Extensive experiments and comparison with existing state of the art algorithms show the effectiveness of the proposed algorithm.

MixNet for Generalized Face Presentation Attack Detection

Abstract: The non-intrusive nature and high accuracy of face recognition algorithms have led to their successful deployment across multiple applications ranging from border access to mobile unlocking and digital payments. However, their vulnerability against sophisticated and cost-effective presentation attack mediums raises essential questions regarding its reliability. In the literature, several presentation attack detection algorithms are presented; however, they are still far behind from reality. The major problem with existing work is the generalizability against multiple attacks both in the seen and unseen setting. The algorithms which are useful for one kind of attack (such as print) perform unsatisfactorily for another type of attack (such as silicone masks). In this research, we have proposed a deep learning-based network termed as MixNet to detect presentation attacks in cross-database and unseen attack settings. The proposed algorithm utilizes state-of-the-art convolutional neural network architectures and learns the feature mapping for each attack category. Experiments are performed using multiple challenging face presentation attack databases such as SMAD and Spoof In the Wild (SiW-M) databases. Extensive experiments and comparison with existing state of the art algorithms show the effectiveness of the proposed algorithm.

Generalized Contact Lens Iris Presentation Attack Detection

Abstract: The high accuracy of iris recognition for person identification has led to its deployment for a variety of applications ranging from border access to mobile unlocking to digital payment. In addition, the commercial success of mobile devices for iris image acquisition enables the easy acquisition of iris images both in an indoor controlled environment as well as an uncontrolled outdoor environment. At the same time, iris recognition systems can easily be attacked using wearable contact lenses. In the literature, several contact lens detection algorithms are proposed; however, the significant drawback is the generalizability under unseen testing domain images. In this research, a novel 3D contact lens iris presentation attack detection algorithm is developed and extensive experiments are performed. The experiments are performed using multiple challenging iris presentation attack databases including the IIITD and LivDet. For the evaluation, we have utilized the experimental protocols, which reflect in-the-wild settings for 3D contact lens iris presentation attack detection where the images belong to both controlled and adverse imaging conditions. The comparison with several state-of-the-art algorithms establishes the effectiveness of the proposed algorithm.

Unified Detection of Digital and Physical Face Attacks.

Abstract: State-of-the-art defense mechanisms against face attacks achieve near perfect accuracies within one of three attack categories, namely adversarial, digital manipulation, or physical spoofs, however, they fail to generalize well when tested across all three categories Poor generalization can be attributed to learning incoherent attacks jointly To overcome this shortcoming, we propose a unified attack detection framework, namely UniFAD, that can automatically cluster 25 coherent attack types belonging to the three categories Using a multi-task learning framework along with k-means clustering, UniFAD learns joint representations for coherent attacks, while uncorrelated attack types are learned separately Proposed UniFAD outperforms prevailing defense methods and their fusion with an overall TDR = 9473% @ 02% FDR on a large fake face dataset consisting of 341K bona fide images and 448K attack images of 25 types across all 3 categories Proposed method can detect an attack within 3 milliseconds on a Nvidia 2080Ti UniFAD can also identify the attack types and categories with 7581% and 9737% accuracies, respectively

Boosting Face Presentation Attack Detection in Multi-Spectral Videos Through Score Fusion of Wavelet Partition Images

Abstract: Presentation attack detection (PAD) algorithms have become an integral requirement for the secure usage of face recognition systems. As face recognition algorithms and applications increase from constrained to unconstrained environments and in multispectral scenarios, presentation attack detection algorithms must also increase their scope and effectiveness. It is important to realize that the PAD algorithms are not only effective for one environment or condition but rather be generalizable to a multitude of variabilities that are presented to a face recognition algorithm. With this motivation, as the first contribution, the article presents a unified PAD algorithm for different kinds of attacks such as printed photos, a replay of video, 3D masks, silicone masks, and wax faces. The proposed algorithm utilizes a combination of wavelet decomposed raw input images from sensor and face region data to detect whether the input image is bonafide or attacked. The second contribution of the article is the collection of a large presentation attack database in the NIR spectrum, containing images from individuals of two ethnicities. The database contains 500 print attack videos which comprise approximately 1,00,000 frames collectively in the NIR spectrum. Extensive evaluation of the algorithm on NIR images as well as visible spectrum images obtained from existing benchmark databases shows that the proposed algorithm yields state-of-the-art results and surpassed several complex and state-of-the-art algorithms. For instance, on benchmark datasets, namely CASIA-FASD, Replay-Attack, and MSU-MFSD, the proposed algorithm achieves a maximum error of 0.92% which is significantly lower than state-of-the-art attack detection algorithms.

Multi-stream Face Anti-spoofing System Using 3D Information

Abstract: Face anti-spoofing plays a crucial role in face recog-nition systems widely used in smart devices and security systems. In this paper, we propose a multi-stream fusion system based on a 3D camera by making full use of 3D information for face anti-spoofing. This 3D information is composed of depth maps and Surface Normal Maps (SNM). Detailed discussions about systems are given. Comparison among different modalities and comparison among other methods are provided through several experiments on the public WMCA dataset and our self-build Anti-3D dataset. Due to fewer parameters and less time overhead, we also implement the system on hardware platforms with a 3D camera.