We propose a classification technique for face expression recognition using AdaBoost that learns by selecting the relevant global and local appearance features with the most discriminating information. Selectivity reduces the dimensionality of the feature space that in turn results in significant speed up during online classification. We compare our method with another leading margin-based classifier, the Support Vector Machines (SVM) and identify the advantages of using AdaBoost over SVM in this context. We use histograms of Gabor and Gaussian derivative responses as the appearance features. We apply our approach to the face expression recognition problem where local appearances play an important role. Finally, we show that though SVM performs equally well, AdaBoost feature selection provides a final hypothesis model that can easily be visualized and interpreted, which is lacking in the high dimensional support vectors of the SVM.
Abstract. In recent years, distributed computer vision has gained a lot of attention within the computer vision community for applications such as video surveillance and object tracking. The collective information gathered by multiple cameras that are strategically placed has many advantages. For example, aggregation of information from multiple viewpoints reduces the uncertainty about the scene. Further, there is no single point of failure, thus the system as a whole could continue to perform the task at hand. However, the advantages arising out of such cooperation can be realized only by timely sharing of the information between them. This paper discusses the design of a distributed vision system that enables several heterogeneous sensors with different processing rates to exchange information in a timely manner in order to achieve a common goal, say tracking of multiple human subjects and mobile robots in an indoor smart environment. In our fault-tolerant distributed vision system, a resource manager manages individual cameras and buffers the time-stamped object candidates received from them. A User Agent with a given task specification approaches the resource manager, first for knowing the available resources (cameras) and later for receiving the object candidates from the resources of its interest. Thus the resource manager acts as a proxy between the user agents and cameras, thereby freeing the cameras to do dedicated feature detection and extraction only. In such a scenario, many failures are possible. For example, one of the cameras may have a hardware failure or it may lose the target, which moved away from its field of view. In this context, important issues such as failure detection and handling, synchronization of data from multiple sensors and sensor reconfiguration by view planning are discussed in the paper. Experimental results with real scene images will be given.
Abstract-In robot and sensor networks, one of the key challenges is to decide when and where to deploy sensory resources to gather information of optimal value. The problem is essentially one of planning, scheduling and controlling the sensors in the network to acquire data from an environment that is constantly varying. The dynamic nature of the problem precludes the use of traditional rule-based strategies that can handle only quasi-static context changes. Automatic context derivation procedures are thus essential for providing fault recovery and fault pre-emption in such systems.We posit that the quality of a sensor network configuration depends on sensor coverage and geometry, sensor allocation policies, and the dynamic processes in the environment. In this paper, we show how these factors can be manipulated in an adaptive framework for robust run-time resource management. We demonstrate our ideas in a people tracking application using a network of multiple cameras. The task specification for our multi-camera network is one of allocating a camera pair that can best localize a human subject given the current context. The system automatically derives policies for switching between camera pairs that enable robust tracking while being attentive to performance measures. Our approach is unique in that we do not make any a priori assumptions about the scene or the activities that take place in the scene. Models of motion dynamics in the scene and the camera network configuration steer the policies to provide robust tracking.
AbsOacI-In this paper we propose a new approach for multiobjective contml nslng a cascade of filters that PmgmsivelY Lpmoves candidate command5 which do not satisfy lask constraints.complex conflicts between objectives that the control system must resolve.The approach is motivated by other contml methods that prevent destructive wntrol interactions through null space projections. We apply this appmach to a practical IeaderKollower task in which a mobile robot must address the conflicting objectives of moving to a goal position while avoiding obstacles and keeping a region of the workspace within the field of view of a fixed camera mounted on the platform. We experimentslly verify our approach using the Segway Robotic Mobility Platform (RMP), a dynamically stable, differential drive mobile robot.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.