Sumantra Dutta Roy

Active recognition through next view planning: a survey

Abstract 3-D object recognition involves using image-computable features to identify 3-D object. A single view of a 3-D object may not contain sufficient features to recognize it unambiguously. One needs to plan different views around the given object in order to recognize it. Such a task involves an active sensor—one whose parameters (external and/or internal) can be changed in a purposive manner. In this paper, we review two important applications of an active sensor. We first survey important approaches to active 3-D object recognition. Next, we review existing approaches towards another important application of an active sensor namely, that of scene analysis and interpretation.

Hand gesture modelling and recognition involving changing shapes and trajectories, using a Predictive EigenTracker

We present a novel eigenspace-based framework to model a dynamic hand gesture that incorporates both hand shape as well as trajectory information. We address the problem of choosing a gesture set that models an upper bound on gesture recognition efficiency. We show encouraging experimental results on a such a representative set.

Automatic classification and prediction models for early Parkinson's disease diagnosis from SPECT imaging

Propose methods for very accurate classification of early PD using only 4 features.Used public database which is large and diverse making the developed models robust.First study to develop accurate prognostic model based on SBR features for early PD. Early and accurate diagnosis of Parkinsons disease (PD) is important for early management, proper prognostication and for initiating neuroprotective therapies once they become available. Recent neuroimaging techniques such as dopaminergic imaging using single photon emission computed tomography (SPECT) with 123I-Ioflupane (DaTSCAN) have shown to detect even early stages of the disease. In this paper, we use the striatal binding ratio (SBR) values that are calculated from the 123I-Ioflupane SPECT scans (as obtained from the Parkinsons progression markers initiative (PPMI) database) for developing automatic classification and prediction/prognostic models for early PD. We used support vector machine (SVM) and logistic regression in the model building process. We observe that the SVM classifier with RBF kernel produced a high accuracy of more than 96% in classifying subjects into early PD and healthy normal; and the logistic model for estimating the risk of PD also produced high degree of fitting with statistical significance indicating its usefulness in PD risk estimation. Hence, we infer that such models have the potential to aid the clinicians in the PD diagnostic process.

Isolated 3D object recognition through next view planning

In many cases, a single view of an object may not contain sufficient features to recognize it unambiguously. This paper presents a new online recognition scheme based on next view planning for the identification of an isolated 3D object using simple features. The scheme uses a probabilistic reasoning framework for recognition and planning. Our knowledge representation scheme encodes feature based information about objects as well as the uncertainty in the recognition process. This is used both in the probability calculations as well as in planning the next view. Results clearly demonstrate the effectiveness of our strategy for a reasonably complex experimental set.

A Fast Method for Image Mosaicing using Geometric Hashing

The general problem of mosaicing is to create a single seamless image by aligning a series of spatially overlapped images. The result is an image with a field of view greater than that of a single image. Traditionally this research has been aimed at stitching together images taken by aerial or satellite reconnaissance equipment. With the advancement of personal computing equipment, the creation of image mosaic has entered the consumer market. Thus, automation of the process is an important issue. This paper proposes a new method for automatic generation of mosaics using Geometric Hashing. This speeds up the matching process. We show the application of our method on two important cases namely, rigid planar motion and panoramic mosaics. We provide experimental results in support of our proposed method.

Developing a Gesture-based Interface

A gesture-based interface involves tracking a moving hand across frames, and extracting the semantic interpretation corresponding to the gesture. This is a difficult task, since there is a change in both the position as well as the appearance of the hand. Further, such a system should be robust to the speed at which the gesture is performed. This paper presents a novel attempt at developing a hand gesture-based interface. We propose an on-line predictive EigenTracker for the moving hand. Our tracker can learn the eigenspace on the fly. We propose a new state-based representation scheme for hand gestures, based on the eigenspace reconstruction error. This makes the system independent of the speed of performing the gesture. We use learning for adapting the gesture recognition system to individual requirements. We show results of successful operation of our system even in cases of background clutter and other moving objects.

Robot localization using uncalibrated camera invariants

We describe a set of image measurements which are invariant to the camera internals but are location variant. We show that using these measurements it is possible to calculate the self-localization of a robot using known landmarks and uncalibrated cameras. We also show that it is possible to compute, using uncalibrated cameras, the Euclidean structure of 3-D world points using multiple views from known positions. We are free to alter the internal parameters of the camera during these operations. Our initial experiments demonstrate the applicability of the method.

Recognizing large 3-D objects through next view planning using an uncalibrated camera

We present a new on-line scheme for the recognition and pose estimation of a large isolated 3-D object, which may not entirely fit in a cameras field of view. We do not assume any knowledge of the internal parameters of the camera, or their constancy. We use a probabilistic reasoning framework for recognition and next view planning. We show results of successful recognition and pose estimation even in cases of a high degree of interpretation ambiguity associated with the initial view.

High-Accuracy Detection of Early Parkinson's Disease through Multimodal Features and Machine Learning

Early (or preclinical) diagnosis of Parkinsons disease (PD) is crucial for its early management as by the time manifestation of clinical symptoms occur, more than 60% of the dopaminergic neurons have already been lost. It is now established that there exists a premotor stage, before the start of these classic motor symptoms, characterized by a constellation of clinical features, mostly non-motor in nature such as Rapid Eye Movement (REM) sleep Behaviour Disorder (RBD) and olfactory loss. In this paper, we use the non-motor features of RBD and olfactory loss, along with other significant biomarkers such as Cerebrospinal fluid (CSF) measurements and dopaminergic imaging markers from 183 healthy normal and 401 early PD subjects, as obtained from the Parkinsons Progression Markers Initiative (PPMI) database, to classify early PD subjects from normal using Naïve Bayes, Support Vector Machine (SVM), Boosted Trees and Random Forests classifiers. We observe that SVM classifier gave the best performance (96.40% accuracy, 97.03% sensitivity, 95.01% specificity, and 98.88% area under ROC). We infer from the study that a combination of non-motor, CSF and imaging markers may aid in the preclinical diagnosis of PD.

Recognizing large isolated 3-D objects through next view planning using inner camera invariants

Most model-based three-dimensional (3D) object recognition systems use information from a single view of an object. However, a single view may not contain sufficient features to recognize it unambiguously. Further, two objects may have all views in common with respect to a given feature set, and may be distinguished only through a sequence of views. A further complication arises when in an image, we do not have a complete view of an object. This paper presents a new online scheme for the recognition and pose estimation of a large isolated 3D object, which may not entirely fit in a cameras field of view. We consider an uncalibrated projective camera, and consider the case when the internal parameters of the camera may be varied either unintentionally, or on purpose. The scheme uses a probabilistic reasoning framework for recognition and next-view planning. We show results of successful recognition and pose estimation even in cases of a high degree of interpretation ambiguity associated with the initial view.