A human pose often conveys not only the configuration of the body parts, but also implicit predictive information about the ensuing motion. This dynamic information can benefit vision applications which lack explicit motion cues. The human visual system can easily perceive the dynamic information in still images. However, computational algorithms to infer and utilize it in computer vision applications are limited. In this paper, we propose a probabilistic framework to infer the dynamic information associated with a human pose. The inference problem is posed as a nonparametric density estimation problem on a non-Euclidean manifold of linear dynamical models. Since direct modeling is intractable, we develop a data driven approach, estimating the density for the test sample under consideration. Statistical inference on the estimated density provides us with quantities of interest like the most probable future motion of the human and the amount of motion information conveyed by a pose. Our experiments demonstrate that the extracted motion information benefits numerous applications in computer vision. In particular, the predicted future motion is useful for activity recognition, human trajectory synthesis, and motion prediction.

Iris biometric entails using the patterns on the iris as a biometric for personal authentication. It has additional benefits over contact-based biometrics such as fingerprints and hand geometry. However, iris biometric often suffers from the following three challenges: ability to handle unconstrained acquisition, privacy enhancement without compromising security, and robust matching. This chapter discusses a unified framework based on sparse representations and random projections that can address these issues simultaneously. Furthermore, recognition from iris videos as well as generation of cancelable iris templates for enhancing the privacy and security is also discussed.

We present a multi-class multiple instance learning (MIL) algorithm using the dictionary learning framework where the data is given in the form of bags. Each bag contains multiple samples, called instances, out of which at least one belongs to the class of the bag. We propose a noisy-OR model and a generalized mean-based optimization framework for learning the dictionaries in the feature space. The proposed method can be viewed as a generalized dictionary learning algorithm since it reduces to a novel discriminative dictionary learning framework when there is only one instance in each bag. Various experiments using popular vision-related MIL datasets as well as the UNBC-McMaster Pain Shoulder Archive database show that the proposed method performs significantly better than the existing methods.

In this paper, we develop a multiple instance learning (MIL) algorithm using the dictionary learning framework where the labels are given in the form of positive and negative bags, with each bag containing multiple samples. A positive bag is guaranteed to have only one positive class sample while all the samples in a negative bag belong to the negative class. Given positive and negative bags of data, our method learns appropriate feature space to select positive samples from the positive bags as well as optimal dictionaries to represent data in these bags. We apply this method for digit recognition, action recognition, and gender recognition tasks and demonstrate that the proposed method is robust and can perform significantly better than many competitive two class MIL classification algorithms.

We present a multi-class, multiple instance learning (MIL) algorithm using the dictionary learning framework where the data is given in the form of bags. Each bag contains multiple samples, called instances, out of which at least one belongs to the class of the bag. We propose a noisy-OR model-based optimization framework for learning the dictionaries. Our method can be viewed as a generalized dictionary learning algorithm since it reduces to a novel discriminative dictionary learning framework when there is only one instance in each bag. Various experiments using the popular MIL datasets show that the proposed method performs better than existing methods.

Due to the increasing popularity of iris biometrics, new sensors are being developed for acquiring iris images and existing ones are being continuously upgraded. Re-enrolling users every time a new sensor is deployed is expensive and time-consuming, especially in applications with a large number of enrolled users. However, recent studies show that cross-sensor matching, where the test samples are verified using data enrolled with a different sensor, often lead to reduced performance. In this paper, we propose a machine learning technique to mitigate the cross-sensor performance degradation by adapting the iris samples from one sensor to another. We first present a novel optimization framework for learning transformations on iris biometrics. We then utilize this framework for sensor adaptation, by reducing the distance between samples of the same class, and increasing it between samples of different classes, irrespective of the sensors acquiring them. Extensive evaluations on iris data from multiple sensors demonstrate that the proposed method leads to improvement in cross-sensor recognition accuracy. Furthermore, since the proposed technique requires minimal changes to the iris recognition pipeline, it can easily be incorporated into existing iris recognition systems.

PROBLEM TO BE SOLVED: To improve sorting of pixel levels of a scene and a possible layout while using novel geometrical restrictions and associated optimization procedures.SOLUTION: A three-dimensional (3D) real-world scene is reconstituted from a single two-dimensional (2D) image by identifying a coupling point which satisfies geometrical restrictions of the scene on the basis of cross lines which are orthogonal with each other, a disappearance point and a disappearance line. By sampling the 2D image in accordance with the coupling point, a possible layout of the scene is generated. Next, an energy function is maximized so as to select an optimal layout from among possible layouts. The energy function evaluates the possible layout while using a conditional random field (CRF) model.

Junctions are strong cues for understanding the geometry of a scene. In this paper, we consider the problem of detecting junctions and using them for recovering the spatial layout of an indoor scene. Junction detection has always been challenging due to missing and spurious lines. We work in a constrained Manhattan world setting where the junctions are formed by only line segments along the three principal orthogonal directions. Junctions can be classified into several categories based on the number and orientations of the incident line segments. We provide a simple and efficient voting scheme to detect and classify these junctions in real images. Indoor scenes are typically modeled as cuboids and we formulate the problem of the cuboid layout estimation as an inference problem in a conditional random field. Our formulation allows the incorporation of junction features and the training is done using structured prediction techniques. We outperform other single view geometry estimation methods on standard datasets.

In computer vision applications, features often lie on Riemannian manifolds with known geometry. Popular learning algorithms such as discriminant analysis, partial least squares, support vector machines, etc., are not directly applicable to such features due to the non-Euclidean nature of the underlying spaces. Hence, classification is often performed in an extrinsic manner by mapping the manifolds to Euclidean spaces using kernels. However, for kernel based approaches, poor choice of kernel often results in reduced performance. In this paper, we address the issue of kernel selection for the classification of features that lie on Riemannian manifolds using the kernel learning approach. We propose two criteria for jointly learning the kernel and the classifier using a single optimization problem. Specifically, for the SVM classifier, we formulate the problem of learning a good kernel-classifier combination as a convex optimization problem and solve it efficiently following the multiple kernel learning approach. Experimental results on image set-based classification and activity recognition clearly demonstrate the superiority of the proposed approach over existing methods for classification of manifold features.

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.

While recent techniques for discriminative dictionary learning have demonstrated tremendous success in image analysis applications, their performance is often limited by the amount of labeled data available for training. Even though labeling images is difficult, it is relatively easy to collect unlabeled images either by querying the web or from public datasets. In this paper, we propose a discriminative dictionary learning technique which utilizes both labeled and unlabeled data for learning dictionaries. Extensive evaluation on existing datasets demonstrate that the proposed method performs significantly better than state of the art dictionary learning approaches when unlabeled images are available for training.

The labeling of large sets of images for training or testing analysis systems can be a very costly and time-consuming process. Multiple instance learning (MIL) is a generalization of traditional supervised learning which relaxes the need for exact labels on training instances. Instead, the labels are required only for a set of instances known as bags. In this paper, we apply MIL to the retrieval and localization of signatures and the retrieval of images containing machine-printed text, and show that a gain of 15-20% in performance can be achieved over the supervised learning with weak-labeling. We also compare our approach to supervised learning with fully annotated training data and report a competitive accuracy for MIL. Using our experiments on real-world datasets, we show that MIL is a good alternative when the training data has only document-level annotation.

Noncontact biometrics such as face and iris have additional benefits over contact-based biometrics such as fingerprint and hand geometry. However, three important challenges need to be addressed in a noncontact biometrics-based authentication system: ability to handle unconstrained acquisition, robust and accurate matching, and privacy enhancement without compromising security. In this paper, we propose a unified framework based on random projections and sparse representations, that can simultaneously address all three issues mentioned above in relation to iris biometrics. Our proposed quality measure can handle segmentation errors and a wide variety of possible artifacts during iris acquisition. We demonstrate how the proposed approach can be easily extended to handle alignment variations and recognition from iris videos, resulting in a robust and accurate system. The proposed approach includes enhancements to privacy and security by providing ways to create cancelable iris templates. Results on public data sets show significant benefits of the proposed approach.

Privacy and security are essential requirements in practical biometric systems. In order to prevent the theft of biometric patterns, it is desired to modify them through revocable and non invertible transformations called Cancelable Biometrics. In this paper, we propose an efficient algorithm for generating a Cancelable Iris Biometric based on Sectored Random Projections. Our algorithm can generate a new pattern if the existing one is stolen, retain the original recognition performance and prevent extraction of useful information from the transformed patterns. Our method also addresses some of the drawbacks of existing techniques and is robust to degradations due to eyelids and eyelashes.

Iris images acquired from a partially cooperating subject often suffer from blur, occlusion due to eyelids, and specular reflections. The performance of existing iris recognition systems degrade significantly on these images. Hence it is essential to select good images from the incoming iris video stream, before they are input to the recognition algorithm. In this paper, we propose a sparsity based algorithm for selection of good iris images and their subsequent recognition. Unlike most existing algorithms for iris image selection, our method can handle segmentation errors and a wider range of acquisition artifacts common in iris image capture. We perform selection and recognition in a single step which is more efficient than devising separate specialized algorithms for the two. Recognition from partially cooperating users is a significant step towards deploying iris systems in a wide variety of applications.