Rotation Invariant Object Recognition from One Training Example

Local descriptors are increasingly used for the task of object recognition because of their perceived robustness with respect to occlusions and to global geometrical deformations. Such a descriptor--based on a set of oriented Gaussian derivative filters-- is used in our recognition system. We report here an evaluation of several techniques for orientation estimation to achieve rotation invariance of the descriptor. We also describe feature selection based on a single training image. Virtual images are generated by rotating and rescaling the image and robust features are selected. The results confirm robust performance in cluttered scenes, in the presence of partial occlusions, and when the object is embedded in different backgrounds.

Support Vector Machines: Training and Applications

The Support Vector Machine (SVM) is a new and very promising classification technique developed by Vapnik and his group at AT&T Bell Labs. This new learning algorithm can be seen as an alternative training technique for Polynomial, Radial Basis Function and Multi-Layer Perceptron classifiers. An interesting property of this approach is that it is an approximate implementation of the Structural Risk Minimization (SRM) induction principle. The derivation of Support Vector Machines, its relationship with SRM, and its geometrical insight, are discussed in this paper. Training a SVM is equivalent to solve a quadratic programming problem with linear and box constraints in a number of variables equal to the number of data points. When the number of data points exceeds few thousands the problem is very challenging, because the quadratic form is completely dense, so the memory needed to store the problem grows with the square of the number of data points. Therefore, training problems arising in some real applications with large data sets are impossible to load into memory, and cannot be solved using standard non-linear constrained optimization algorithms. We present a decomposition algorithm that can be used to train SVM's over large data sets. The main idea behind the decomposition is the iterative solution of sub-problems and the evaluation of, and also establish the stopping criteria for the algorithm. We present previous approaches, as well as results and important details of our implementation of the algorithm using a second-order variant of the Reduced Gradient Method as the solver of the sub-problems. As an application of SVM's, we present preliminary results we obtained applying SVM to the problem of detecting frontal human faces in real images.

Workforce Issues in the Greater Boston Health Care Industry: Implications for Work and Family

Interviews with more than 40 leaders in the Boston area health care industry have identified a range of broadly-felt critical problems. This document synthesizes these problems and places them in the context of work and family issues implicit in the organization of health care workplaces. It concludes with questions about possible ways to address such issues. The defining circumstance for the health care industry nationally as well as regionally at present is an extraordinary reorganization, not yet fully negotiated, in the provision and financing of health care. Hoped-for controls on increased costs of medical care – specifically the widespread replacement of indemnity insurance by market-based managed care and business models of operation--have fallen far short of their promise. Pressures to limit expenditures have produced dispiriting conditions for the entire healthcare workforce, from technicians and aides to nurses and physicians. Under such strains, relations between managers and workers providing care are uneasy, ranging from determined efforts to maintain respectful cooperation to adversarial negotiation. Taken together, the interviews identify five key issues affecting a broad cross-section of occupational groups, albeit in different ways: Staffing shortages of various kinds throughout the health care workforce create problems for managers and workers and also for the quality of patient care. Long work hours and inflexible schedules place pressure on virtually every part of the healthcare workforce, including physicians. Degraded and unsupportive working conditions, often the result of workplace "deskilling" and "speed up," undercut previous modes of clinical practice. Lack of opportunities for training and advancement exacerbate workforce problems in an industry where occupational categories and terms of work are in a constant state of flux. Professional and employee voices are insufficiently heard in conditions of rapid institutional reorganization and consolidation. Interviewees describe multiple impacts of these issues--on the operation of health care workplaces, on the well being of the health care workforce, and on the quality of patient care. Also apparent in the interviews, but not clearly named and defined, is the impact of these issues on the ability of workers to attend well to the needs of their families--and the reciprocal impact of workers' family tensions on workplace performance. In other words, the same things that affect patient care also affect families, and vice versa. Some workers describe feeling both guilty about raising their own family issues when their patients' needs are at stake, and resentful about the exploitation of these feelings by administrators making workplace policy. The different institutions making up the health care system have responded to their most pressing issues with a variety of specific stratagems but few that address the complexities connecting relations between work and family. The MIT Workplace Center proposes a collaborative exploration of next steps to probe these complications and to identify possible locations within the health care system for workplace experimentation with outcomes benefiting all parties.

Validating Co-Training Models for Web Image Classification

Co-training is a semi-supervised learning method that is designed to take advantage of the redundancy that is present when the object to be identified has multiple descriptions. Co-training is known to work well when the multiple descriptions are conditional independent given the class of the object. The presence of multiple descriptions of objects in the form of text, images, audio and video in multimedia applications appears to provide redundancy in the form that may be suitable for co-training. In this paper, we investigate the suitability of utilizing text and image data from the Web for co-training. We perform measurements to find indications of conditional independence in the texts and images obtained from the Web. Our measurements suggest that conditional independence is likely to be present in the data. Our experiments, within a relevance feedback framework to test whether a method that exploits the conditional independence outperforms methods that do not, also indicate that better performance can indeed be obtained by designing algorithms that exploit this form of the redundancy when it is present.