Miriam Cha

Robust local binary pattern feature sets for periocular biometric identification

In this paper, we perform a detailed investigation of various features that can be extracted from the periocular region of human faces for biometric identification. The emphasis of this study is to explore the BEST feature extraction approach used in stand-alone mode without any generative or discriminative subspace training. Simple distance measures are used to determine the verification rate (VR) on a very large dataset. Several filter-based techniques and local feature extraction methods are explored in this study, where we show an increase of 15% verification performance at 0.1% false accept rate (FAR) compared to raw pixels with the proposed Local Walsh-Transform Binary Pattern encoding. Additionally, when fusing our best feature extraction method with Kernel Correlation Feature Analysis (KCFA) [36], we were able to obtain VR of 61.2%. Our experiments are carried out on the large validation set of the NIST FRGC database [6], which contains facial images from environments with uncontrolled illumination. Verification experiments based on a pure 1–1 similarity matrix of 16028×8014 (~128 million comparisons) carried out on the entire database, where we find that we can achieve a raw VR of 17.0% at 0.1% FAR using our proposed Local Walsh-Transform Binary Pattern approach. This result, while may seem low, is more than the NIST reported baseline VR on the same dataset (12% at 0.1% FAR), when PCA was trained on the entire facial features for recognition [6].

Detecting Depression using Vocal, Facial and Semantic Communication Cues

Major depressive disorder (MDD) is known to result in neurophysiological and neurocognitive changes that affect control of motor, linguistic, and cognitive functions. MDDs impact on these processes is reflected in an individuals communication via coupled mechanisms: vocal articulation, facial gesturing and choice of content to convey in a dialogue. In particular, MDD-induced neurophysiological changes are associated with a decline in dynamics and coordination of speech and facial motor control, while neurocognitive changes influence dialogue semantics. In this paper, biomarkers are derived from all of these modalities, drawing first from previously developed neurophysiologically-motivated speech and facial coordination and timing features. In addition, a novel indicator of lower vocal tract constriction in articulation is incorporated that relates to vocal projection. Semantic features are analyzed for subject/avatar dialogue content using a sparse coded lexical embedding space, and for contextual clues related to the subjects present or past depression status. The features and depression classification system were developed for the 6th International Audio/Video Emotion Challenge (AVEC), which provides data consisting of audio, video-based facial action units, and transcribed text of individuals communicating with the human-controlled avatar. A clinical Patient Health Questionnaire (PHQ) score and binary depression decision are provided for each participant. PHQ predictions were obtained by fusing outputs from a Gaussian staircase regressor for each feature set, with results on the development set of mean F1=0.81, RMSE=5.31, and MAE=3.34. These compare favorably to the challenge baseline development results of mean F1=0.73, RMSE=6.62, and MAE=5.52. On test set evaluation, our system obtained a mean F1=0.70, which is similar to the challenge baseline test result. Future work calls for consideration of joint feature analyses across modalities in an effort to detect neurological disorders based on the interplay of motor, linguistic, affective, and cognitive components of communication.

Test statistics for synthetic aperture radar coherent change detection

Coherent change detection using paired synthetic aperture radar images is typically performed using a classical estimator of coherence applied under an assumption of complex Gaussian data. The magnitudes of the resultant coherence estimates are plotted as an image and used to gauge changes in the observed scene. Here we investigate the suitability of an alternative coherence estimator that further assumes the variances of the populations underlying each paired sample to be equal. We show experimentally that this alternative estimator outperforms the classical estimator even when the underlying variances are not equal, as long as they are close enough. We demonstrate the suitability of this estimator directly on publicly available synthetic aperture radar data, with a performance improvement observed through increased contrast in the corresponding coherent change detection images.

Finding curves in SAR CCD images

This paper introduces a pattern recognition and computer vision approach to mitigating false alarms in synthetic aperture radar (SAR) coherence change detection (CCD) images. In this paper, we perform an automatic detection of roads in SAR CCD images. The approach is based on a curve tracing algorithm originally proposed by Steger with modifications to better suit the goal of curve detection in SAR CCD images [1]. In our technique, the traditional Stegers method is used to detect curve points, and cubic splines are used to approximate the original curve. To detect roads more accurately, preprocessing and outlier removal techniques are performed along with the curve detection.

Two-Stage Change Detection for Synthetic Aperture Radar

Coherent change detection using paired synthetic aperture radar (SAR) images is often performed using a classical coherence estimator that is invariant to the true variances of the populations underlying each paired sample. While attractive, this estimator is biased and requires a significant number of samples to yield good performance. Increasing sample size often results in decreased image resolution. Thus, we propose the use of Bergers coherence estimate because, with the same number of pixels, the estimator effectively doubles the sample support without sacrificing resolution when the underlying population variances are equal or near equal. A potential drawback of this approach is that it is not invariant since its distribution depends on the pixel pair population variances. While Bergers estimator is inherently sensitive to the inequality of population variances, we propose a method of insulating the detector from this acuity. A two-stage change statistic is introduced to combine a noncoherent intensity change statistic given by the sample variance ratio, followed by the alternative Berger estimator, which assumes equal population variances. The first-stage detector identifies pixel pairs that have nonequal variances as changes caused by the displacement of sizeable object. The pixel pairs that are identified to have equal or near-equal variances in the first stage are used as an input to the second stage. The second-stage test uses the alternative Berger coherence estimator to detect subtle changes such as tire tracks and footprints. We show experimentally that the proposed method yields higher contrast SAR change detection images than the classical coherent change detector (state of the art), the alternative coherent change detector, and the intensity change detector. Experimental results are presented to show the effectiveness and robustness of the proposed algorithm for SAR change detection.

Deep Sparse-coded Network (DSN)

We present Deep Sparse-coded Network (DSN), a deep architecture based on multilayer sparse coding. It has been considered difficult to learn a useful feature hierarchy by stacking sparse coding layers in a straightforward manner. The primary reason is the modeling assumption for sparse coding that takes in a dense input and yields a sparse output vector. Applying a sparse coding layer on the output of another tends to violate the modeling assumption. We overcome this shortcoming by interlacing nonlinear pooling units. Average- or max-pooled sparse codes are aggregated to form dense input vectors for the next sparse coding layer. Pooling achieves nonlinear activation analogous to neural networks while not introducing diminished gradient flows during the training. We introduce a novel backpropagation algorithm to finetune the proposed DSN beyond the pretraining via greedy layerwise sparse coding and dictionary learning. We build an experimental 4-layer DSN with the ℓ1-regularized LARS and the greedy-ℓ0 OMP, and demonstrate superior performance over a similarly-configured stacked autoencoder (SAE) on CIFAR-10.

Joint SAR image compression and coherent change detection

Fine details revealed by synthetic aperture radar (SAR) coherent change detection (CCD), such as foot prints, require SAR imagery with both high resolution and precision. These large data requirements are at odds with the low bandwidths often available for SAR change detection systems such as those that utilize small unmanned aerial vehicles (UAVs). Here we investigate the interplay between SAR data compression and SAR CCD performance. As the data are compressed further, the ability to detect changes decreases. However, there is redundant information contained in SAR imagery that is not necessary for change detection, and removing it makes SAR compression possible. In this paper, we introduce a new model-based compression method that leverages the known distribution of SAR data for a compact storage, while improving change detection performance. We show experimentally that the CCD using the decompressed SAR pair after our proposed method not only yields significant improvement in change detection over the CCD using the decompressed SAR after block adaptive quantization (BAQ) method, but also over the CCD using the original SAR data. Experimental results are presented to show the effectiveness and robustness of the proposed algorithm for SAR compression and change detection.

Automatic track tracing in SAR CCD images using search cues

In this paper, we present an algorithm for automatic vehicle track tracing in synthetic aperture radar coherent change detection (SAR CCD) images using search cues. The framework consists of two main steps. The first step uses a rotating matched filter that is modeled to characterize the appearance of vehicle tracks in SAR CCD imagery. For every pixel, the algorithm searches the orientations of the filter that best match the local orientations of the track path using normalized cross correlation. The second step includes track tracing from the estimated orientation image obtained from the previous step. Given a search cue, the tracing algorithm aims to find a parallel track path that maximizes the global length of the curve, and minimizes the differences in the pixel positions and orientations.

Language Modeling by Clustering with Word Embeddings for Text Readability Assessment

We present a clustering-based language model using word embeddings for text readability prediction. Presumably, an Euclidean semantic space hypothesis holds true for word embeddings whose training is done by observing word co-occurrences. We argue that clustering with word embeddings in the metric space should yield feature representations in a higher semantic space appropriate for text regression. Also, by representing features in terms of histograms, our approach can naturally address documents of varying lengths. An empirical evaluation using the Common Core Standards corpus reveals that the features formed on our clustering-based language model significantly improve the previously known results for the same corpus in readability prediction. We also evaluate the task of sentence matching based on semantic relatedness using the Wiki-SimpleWiki corpus and find that our features lead to superior matching performance.

COMBINED INTENSITY AND COHERENT CHANGE DETECTION FOR SYNTHETIC APERTURE RADAR

Coherent change detection using paired synthetic aperture radar images is performed using a classical coherence estimator applied under an assumption of complex Gaussian data. The magnitudes of the resulting coherence estimates are plotted as an image and used to gauge changes in the observed scene. In this paper, a two-stage change statistic that combines non-coherent and coherent change detection algorithms is proposed. In the first stage, a non-coherent intensity change detector is applied to test for changes caused by the displacement of a sizable object using the sample variance ratio test. The sample pairs that failed the first stage are used as an input to the second stage. The second stage test uses an alternative coherence estimator that assumes equal population variances, to detect subtle changes such as tire tracks and footprints. We show experimentally that the proposed method not only has a superior change detection performance over the classical coherent change detector, but also over either the non-coherent intensity change detector or the alternative coherent change detector, alone. Experimental results are presented to show the effectiveness and robustness of the proposed algorithm for SAR change detection.