A5101771922- Face recognition and analysis
- Face and Expression Recognition
- Biometric Identification and Security
- Video Surveillance and Tracking Methods
- Handwritten Text Recognition Techniques
- Image Retrieval and Classification Techniques
- Image Processing and 3D Reconstruction
- Spam and Phishing Detection
- Network Security and Intrusion Detection
- Adversarial Robustness in Machine Learning
- Advanced Malware Detection Techniques
- Industrial Vision Systems and Defect Detection
- Remote Sensing and Land Use
- User Authentication and Security Systems
- Domain Adaptation and Few-Shot Learning
- Advanced Neural Network Applications
- Advanced Image and Video Retrieval Techniques
- Vehicle License Plate Recognition
University of Maryland, College Park
2016-2019
National Institute of Technology Rourkela
2012-2013
In this paper, automated user verification techniques for smartphones are investigated. A unique non-commercial dataset, the University of Maryland Active Authentication Dataset 02 (UMDAA-02) multi-modal authentication research is introduced. This paper focuses on three sensors - front camera, touch sensor and location service while providing a general description other modalities. Benchmark results face detection, verification, touch-based identification location-based next-place prediction...
In this paper, targeted fooling of high performance image classifiers is achieved by developing two novel attack methods. The first method generates universal perturbations for target classes and the second specific perturbations. Extensive experiments are conducted on MNIST CIFAR10 datasets to provide insights about proposed algorithms show their effectiveness.
State-of-the-art methods of attribute detection from faces almost always assume the presence a full, unoccluded face. Hence, their performance degrades for partially visible and occluded faces. In this paper, we introduce SPLITFACE, deep convolutional neural network-based method that is explicitly designed to perform in Taking several facial segments full face as input, proposed takes data driven approach determine which attributes are localized segments. The unique architecture network...
We propose a deep feature-based face detector for mobile devices to detect user's acquired by the front-facing camera. The proposed method is able faces in images containing extreme pose and illumination variations as well partial faces. main challenge developing algorithms constrained nature of platform non-availability CUDA enabled GPUs on such devices. Our implementation takes into account special captured camera exploits present without CUDA-based frameworks, meet these challenges.
Generic face detection algorithms do not perform very well in the mobile domain due to significant presence of occluded and partially visible faces. One promising technique handle challenge partial faces is design detectors based on facial segments. In this paper two such namely, SegFace DeepSegFace, are proposed that detect a given arbitrary combinations certain Both methods use proposals from segments as input found using weak boosted classifiers. shallow fast algorithm traditional...