Yi-Chen Chen

Bio: Yi-Chen Chen is an academic researcher from University of Maryland, College Park. The author has contributed to research in topics: Sparse approximation & Facial recognition system. The author has an hindex of 9, co-authored 14 publications receiving 429 citations.

Dictionary-based face recognition from video

Abstract: The main challenge in recognizing faces in video is effectively exploiting the multiple frames of a face and the accompanying dynamic signature. One prominent method is based on extracting joint appearance and behavioral features. A second method models a person by temporal correlations of features in a video. Our approach introduces the concept of video-dictionaries for face recognition, which generalizes the work in sparse representation and dictionaries for faces in still images. Video-dictionaries are designed to implicitly encode temporal, pose, and illumination information. We demonstrate our method on the Face and Ocular Challenge Series (FOCS) Video Challenge, which consists of unconstrained video sequences. We show that our method is efficient and performs significantly better than many competitive video-based face recognition algorithms.

Ambiguously Labeled Learning Using Dictionaries

Abstract: We propose a dictionary-based learning method for ambiguously labeled multiclass classification, where each training sample has multiple labels and only one of them is the correct label. The dictionary learning problem is solved using an iterative alternating algorithm. At each iteration of the algorithm, two alternating steps are performed: 1) a confidence update and 2) a dictionary update. The confidence of each sample is defined as the probability distribution on its ambiguous labels. The dictionaries are updated using either soft or hard decision rules. Furthermore, using the kernel methods, we make the dictionary learning framework nonlinear based on the soft decision rule. Extensive evaluations on four unconstrained face recognition datasets demonstrate that the proposed method performs significantly better than state-of-the-art ambiguously labeled learning approaches.

Video-based face recognition via joint sparse representation

Abstract: In video-based face recognition, a key challenge is in exploiting the extra information available in a video; e.g., face, body, and motion identity cues. In addition, different video sequences of the same subject may contain variations in resolution, illumination, pose, and facial expressions. These variations contribute to the challenges in designing an effective video-based face-recognition algorithm. We propose a novel multivariate sparse representation method for video-to-video face recognition. Our method simultaneously takes into account correlations as well as coupling information among the video frames. Our method jointly represents all the video data by a sparse linear combination of training data. In addition, we modify our model so that it is robust in the presence of noise and occlusion. Furthermore, we kernelize the algorithm to handle the non-linearities present in video data. Numerous experiments using unconstrained video sequences show that our method is effective and performs significantly better than many state-of-the-art video-based face recognition algorithms in the literature.

In-Plane Rotation and Scale Invariant Clustering Using Dictionaries

Abstract: In this paper, we present an approach that simultaneously clusters images and learns dictionaries from the clusters. The method learns dictionaries and clusters images in the radon transform domain. The main feature of the proposed approach is that it provides both in-plane rotation and scale invariant clustering, which is useful in numerous applications, including content-based image retrieval (CBIR). We demonstrate the effectiveness of our rotation and scale invariant clustering method on a series of CBIR experiments. Experiments are performed on the Smithsonian isolated leaf, Kimia shape, and Brodatz texture datasets. Our method provides both good retrieval performance and greater robustness compared to standard Gabor-based and three state-of-the-art shape-based methods that have similar objectives.

Dictionary Learning from Ambiguously Labeled Data

Abstract: We propose a novel dictionary-based learning method for ambiguously labeled multiclass classification, where each training sample has multiple labels and only one of them is the correct label. The dictionary learning problem is solved using an iterative alternating algorithm. At each iteration of the algorithm, two alternating steps are performed: a confidence update and a dictionary update. The confidence of each sample is defined as the probability distribution on its ambiguous labels. The dictionaries are updated using either soft (EM-based) or hard decision rules. Extensive evaluations on existing datasets demonstrate that the proposed method performs significantly better than state-of-the-art ambiguously labeled learning approaches.

Dictionary Learning

Abstract: We describe methods for learning dictionaries that are appropriate for the representation of given classes of signals and multisensor data. We further show that dimensionality reduction based on dictionary representation can be extended to address specific tasks such as data analy sis or classification when the learning includes a class separability criteria in the objective function. The benefits of dictionary learning clearly show that a proper understanding of causes underlying the sensed world is key to task-specific representation of relevant information in high-dimensional data sets.

Trunk-Branch Ensemble Convolutional Neural Networks for Video-Based Face Recognition

Abstract: Human faces in surveillance videos often suffer from severe image blur, dramatic pose variations, and occlusion. In this paper, we propose a comprehensive framework based on Convolutional Neural Networks (CNN) to overcome challenges in video-based face recognition (VFR). First, to learn blur-robust face representations, we artificially blur training data composed of clear still images to account for a shortfall in real-world video training data. Using training data composed of both still images and artificially blurred data, CNN is encouraged to learn blur-insensitive features automatically. Second, to enhance robustness of CNN features to pose variations and occlusion, we propose a Trunk-Branch Ensemble CNN model (TBE-CNN), which extracts complementary information from holistic face images and patches cropped around facial components. TBE-CNN is an end-to-end model that extracts features efficiently by sharing the low- and middle-level convolutional layers between the trunk and branch networks. Third, to further promote the discriminative power of the representations learnt by TBE-CNN, we propose an improved triplet loss function. Systematic experiments justify the effectiveness of the proposed techniques. Most impressively, TBE-CNN achieves state-of-the-art performance on three popular video face databases: PaSC, COX Face, and YouTube Faces. With the proposed techniques, we also obtain the first place in the BTAS 2016 Video Person Recognition Evaluation.

Advances in biometrics : International Conference, ICB 2007 Seoul, Korea, August 27-29, 2007 : proceedings

Abstract: Face Recognition.- Super-Resolved Faces for Improved Face Recognition from Surveillance Video.- Face Detection Based on Multi-Block LBP Representation.- Color Face Tensor Factorization and Slicing for Illumination-Robust Recognition.- Robust Real-Time Face Detection Using Face Certainty Map.- Poster I.- Motion Compensation for Face Recognition Based on Active Differential Imaging.- Face Recognition with Local Gabor Textons.- Speaker Verification with Adaptive Spectral Subband Centroids.- Similarity Rank Correlation for Face Recognition Under Unenrolled Pose.- Feature Correlation Filter for Face Recognition.- Face Recognition by Discriminant Analysis with Gabor Tensor Representation.- Fingerprint Enhancement Based on Discrete Cosine Transform.- Biometric Template Classification: A Case Study in Iris Textures.- Protecting Biometric Templates with Image Watermarking Techniques.- Factorial Hidden Markov Models for Gait Recognition.- A Robust Fingerprint Matching Approach: Growing and Fusing of Local Structures.- Automatic Facial Pose Determination of 3D Range Data for Face Model and Expression Identification.- SVDD-Based Illumination Compensation for Face Recognition.- Keypoint Identification and Feature-Based 3D Face Recognition.- Fusion of Near Infrared Face and Iris Biometrics.- Multi-Eigenspace Learning for Video-Based Face Recognition.- Error-Rate Based Biometrics Fusion.- Online Text-Independent Writer Identification Based on Stroke's Probability Distribution Function.- Arm Swing Identification Method with Template Update for Long Term Stability.- Walker Recognition Without Gait Cycle Estimation.- Comparison of Compression Algorithms' Impact on Iris Recognition Accuracy.- Standardization of Face Image Sample Quality.- Blinking-Based Live Face Detection Using Conditional Random Fields.- Singular Points Analysis in Fingerprints Based on Topological Structure and Orientation Field.- Robust 3D Face Recognition from Expression Categorisation.- Fingerprint Recognition Based on Combined Features.- MQI Based Face Recognition Under Uneven Illumination.- Learning Kernel Subspace Classifier.- A New Approach to Fake Finger Detection Based on Skin Elasticity Analysis.- An Algorithm for Biometric Authentication Based on the Model of Non-Stationary Random Processes.- Identity Verification by Using Handprint.- Reducing the Effect of Noise on Human Contour in Gait Recognition.- Partitioning Gait Cycles Adaptive to Fluctuating Periods and Bad Silhouettes.- Repudiation Detection in Handwritten Documents.- A New Forgery Scenario Based on Regaining Dynamics of Signature.- Curvewise DET Confidence Regions and Pointwise EER Confidence Intervals Using Radial Sweep Methodology.- Bayesian Hill-Climbing Attack and Its Application to Signature Verification.- Wolf Attack Probability: A New Security Measure in Biometric Authentication Systems.- Evaluating the Biometric Sample Quality of Handwritten Signatures.- Outdoor Face Recognition Using Enhanced Near Infrared Imaging.- Latent Identity Variables: Biometric Matching Without Explicit Identity Estimation.- Poster II.- 2^N Discretisation of BioPhasor in Cancellable Biometrics.- Probabilistic Random Projections and Speaker Verification.- On Improving Interoperability of Fingerprint Recognition Using Resolution Compensation Based on Sensor Evaluation.- Demographic Classification with Local Binary Patterns.- Distance Measures for Gabor Jets-Based Face Authentication: A Comparative Evaluation.- Fingerprint Matching with an Evolutionary Approach.- Stability Analysis of Constrained Nonlinear Phase Portrait Models of Fingerprint Orientation Images.- Effectiveness of Pen Pressure, Azimuth, and Altitude Features for Online Signature Verification.- Tracking and Recognition of Multiple Faces at Distances.- Face Matching Between Near Infrared and Visible Light Images.- User Classification for Keystroke Dynamics Authentication.- Statistical Texture Analysis-Based Approach for Fake Iris Detection Using Support Vector Machines.- A Novel Null Space-Based Kernel Discriminant Analysis for Face Recognition.- Changeable Face Representations Suitable for Human Recognition.- "3D Face": Biometric Template Protection for 3D Face Recognition.- Quantitative Evaluation of Normalization Techniques of Matching Scores in Multimodal Biometric Systems.- Keystroke Dynamics in a General Setting.- A New Approach to Signature-Based Authentication.- Biometric Fuzzy Extractors Made Practical: A Proposal Based on FingerCodes.- On the Use of Log-Likelihood Ratio Based Model-Specific Score Normalisation in Biometric Authentication.- Predicting Biometric Authentication System Performance Across Different Application Conditions: A Bootstrap Enhanced Parametric Approach.- Selection of Distinguish Points for Class Distribution Preserving Transform for Biometric Template Protection.- Minimizing Spatial Deformation Method for Online Signature Matching.- Pan-Tilt-Zoom Based Iris Image Capturing System for Unconstrained User Environments at a Distance.- Fingerprint Matching with Minutiae Quality Score.- Uniprojective Features for Gait Recognition.- Cascade MR-ASM for Locating Facial Feature Points.- Reconstructing a Whole Face Image from a Partially Damaged or Occluded Image by Multiple Matching.- Robust Hiding of Fingerprint-Biometric Data into Audio Signals.- Correlation-Based Fingerprint Matching with Orientation Field Alignment.- Vitality Detection from Fingerprint Images: A Critical Survey.- Optimum Detection of Multiplicative-Multibit Watermarking for Fingerprint Images.- Fake Finger Detection Based on Thin-Plate Spline Distortion Model.- Robust Extraction of Secret Bits from Minutiae.- Fuzzy Extractors for Minutiae-Based Fingerprint Authentication.- Coarse Iris Classification by Learned Visual Dictionary.- Nonlinear Iris Deformation Correction Based on Gaussian Model.- Shape Analysis of Stroma for Iris Recognition.- Biometric Key Binding: Fuzzy Vault Based on Iris Images.- Multi-scale Local Binary Pattern Histograms for Face Recognition.- Histogram Equalization in SVM Multimodal Person Verification.- Learning Multi-scale Block Local Binary Patterns for Face Recognition.- Horizontal and Vertical 2DPCA Based Discriminant Analysis for Face Verification Using the FRGC Version 2 Database.- Video-Based Face Tracking and Recognition on Updating Twin GMMs.- Poster III.- Fast Algorithm for Iris Detection.- Pyramid Based Interpolation for Face-Video Playback in Audio Visual Recognition.- Face Authentication with Salient Local Features and Static Bayesian Network.- Fake Finger Detection by Finger Color Change Analysis.- Feeling Is Believing: A Secure Template Exchange Protocol.- SVM-Based Selection of Colour Space Experts for Face Authentication.- An Efficient Iris Coding Based on Gauss-Laguerre Wavelets.- Hardening Fingerprint Fuzzy Vault Using Password.- GPU Accelerated 3D Face Registration / Recognition.- Frontal Face Synthesis Based on Multiple Pose-Variant Images for Face Recognition.- Optimal Decision Fusion for a Face Verification System.- Robust 3D Head Tracking and Its Applications.- Multiple Faces Tracking Using Motion Prediction and IPCA in Particle Filters.- An Improved Iris Recognition System Using Feature Extraction Based on Wavelet Maxima Moment Invariants.- Color-Based Iris Verification.- Real-Time Face Detection and Recognition on LEGO Mindstorms NXT Robot.- Speaker and Digit Recognition by Audio-Visual Lip Biometrics.- Modelling Combined Handwriting and Speech Modalities.- A Palmprint Cryptosystem.- On Some Performance Indices for Biometric Identification System.- Automatic Online Signature Verification Using HMMs with User-Dependent Structure.- A Complete Fisher Discriminant Analysis for Based Image Matrix and Its Application to Face Biometrics.- SVM Speaker Verification Using Session Variability Modelling and GMM Supervectors.- 3D Model-Based Face Recognition in Video.- Robust Point-Based Feature Fingerprint Segmentation Algorithm.- Automatic Fingerprints Image Generation Using Evolutionary Algorithm.- Audio Visual Person Authentication by Multiple Nearest Neighbor Classifiers.- Improving Classification with Class-Independent Quality Measures: Q-stack in Face Verification.- Biometric Hashing Based on Genetic Selection and Its Application to On-Line Signatures.- Biometrics Based on Multispectral Skin Texture.- Application of New Qualitative Voicing Time-Frequency Features for Speaker Recognition.- Palmprint Recognition Based on Directional Features and Graph Matching.- Tongue-Print: A Novel Biometrics Pattern.- Embedded Palmprint Recognition System on Mobile Devices.- Template Co-update in Multimodal Biometric Systems.- Continual Retraining of Keystroke Dynamics Based Authenticator.

Unconstrained face verification using deep CNN features

Abstract: In this paper, we present an algorithm for unconstrained face verification based on deep convolutional features and evaluate it on the newly released IARPA Janus Benchmark A (IJB-A) dataset as well as on the traditional Labeled Face in the Wild (LFW) dataset. The IJB-A dataset includes real-world unconstrained faces from 500 subjects with full pose and illumination variations which are much harder than the LFW and Youtube Face (YTF) datasets. The deep convolutional neural network (DCNN) is trained using the CASIA-WebFace dataset. Results of experimental evaluations on the IJB-A and the LFW datasets are provided.