Ramana Isukapalli

Learning a dynamic classification method to detect faces and identify facial expression

While there has been a great deal of research in face detection and recognition, there has been very limited work on identifying the expression on a face. Many current face detection projects use a [Viola/Jones] style “cascade” of Adaboost-based classifiers to interpret (sub)images — e.g. to identify which regions contain faces. We extend this method by learning a decision tree of such classifiers (dtc): While standard cascade classification methods will apply the same sequence of classifiers to each image, our dtc is able to select the most effective classifier at every stage, based on the outcomes of the classifiers already applied. We use dtc not only to detect faces in a test image, but to identify the expression on each face.

WiMAX-EVDO interworking using mobile IP

WiMax, a fourth-generation wireless-access technology, has made significant progress both in the standard forums and with wireless network carriers. To ensure service continuity to a legacy end user, it is important for a WiMax mobile device to interwork with existing third-generation access networks before it is uniquely and ubiquitously deployed. This article addresses this issue and shows how interworking can be achieved with EVDO wireless-access technology, using mobile IP in a dual mode terminal. We present a network architecture solution and detailed call flows.

Global roaming and personal mobility with COPS architecture in SuperDHLR

Personal mobility removes the fixed association between a terminal and a user (a characteristic of traditional fixed and mobile networks), thereby adding one more degree of mobility on top of terminal mobility. Global roaming allows a user to roam in communication networks of different technologies. These two mobility options provide users with ubiquitous services across networks of different types. This paper identifies the technical challenges to achieving global roaming and personal mobility. We propose common operations (COPS) architecture to allow effective multiprotocol support and efficient protocol interworking among disparate networks and compare it with other approaches. SuperDHLR embodies the COPS architecture. It keeps track of user location, manages user profiles for multiple networks, and incorporates service logic for global roaming. It serves as a home location register (HLR) for wireless networks and a mobility management server for Internet protocol (IP) networks. SuperDHLR enables terminal and user mobility and facilitates seamless roaming across circuit and packet switched wireless networks, the Internet and wireline networks.

Learning to detect objects of many classes using binary classifiers

Viola and Jones [VJ] demonstrate that cascade classification methods can successfully detect objects belonging to a single class, such as faces. Detecting and identifying objects that belong to any of a set of “classes”, many class detection, is a much more challenging problem. We show that objects from each class can form a “cluster” in a “classifier space” and illustrate examples of such clusters using images of real world objects. Our detection algorithm uses a “decision tree classifier” (whose internal nodes each correspond to a VJ classifier) to propose a class label for every sub-image W of a test image (or reject it as a negative instance). If this W reaches a leaf of this tree, we then pass W through a subsequent VJ cascade of classifiers, specific to the identified class, to determine whether W is truly an instance of the proposed class. We perform several empirical studies to compare our system for detecting objects of any of M classes, to the obvious approach of running a set of M learned VJ cascade classifiers, one for each class of objects, on the same image. We found that the detection rates are comparable, and our many-class detection system is about as fast as running a single VJ cascade, and scales up well as the number of classes increases.