Promiti Dutta

Evaluation of video browser features and user interaction with VAST MM

In this paper, we present extensive user studies on browsing and information retrieval in the domain of unstructured videos using the VAST MM video library browser. Our studies were performed over a 3-year period with more than 1,000 participants in the university setting. The majority of students use the video library for retrieval of student presentations in a large engineering design course. Through iterative analysis of context-specific audio, visual, and textual cues, we are able to measure significant improvements on typical retrieval tasks, such as searching for unfamiliar content in a large database with over 300 hours of video. We also present user studies conducted in two videotaped core computer science courses to measure the usefulness of the VAST MM (Video Audio Structure Text MultiMedia) resource for final exam preparation. We find that students who use the lecture video library experience significant improvement in final exam scores. To better compare video browsers featuring rich content cues to standard video players without cues, we have performed a large experiment to collect measurable data on search tasks. In general, the lack of index cues can be described by an inverse relationship between amount of matching video content and time required to find it. When index cues are available, the relationship is constant, that is, rare content is found in the same time as common content. We evaluate this data and provide additional insight into two common user interaction techniques: audio-visual browsing and visual-only browsing. We show that user preference is uniform, but that audio-visual browsing is significantly more effective for search and retrieval of video data.

Use of inductive power transfer sharing to increase the driving range of electric vehicles

Electric vehicles have a limited driving range due to current battery capacities. One alternative to larger batteries is sharing charge between vehicles on the road and thereby forming a moving network of vehicles. This is made feasible by the recent advances in inductive power transfer technologies. This paper will demonstrate the feasibility of using this technique through a simulation. We show that with the currently available state of the art technology, our charge sharing method can increase the distance a vehicle travels by more than 50% for a standard driving distribution beyond the standard 100 miles that a vehicle usually travels on one fully charged battery. By adding pricing incentives through game theoretical approaches as well as communication strategies for cars to set rendezvous points, we can ensure participation in our system and symbiotic success for vehicles involved.

Charge sharing model using inductive power transfer to increase feasibility of electric vehicle taxi fleets

Electric vehicles (EVs) are gaining popularity amongst everyday commuters, but are still having a hard time penetrating the fleet vehicle market. In this paper, we investigate a model which does not require changes to current street infrastructure using charge sharing through inductive power transfer to determine if we can extend the driving range of EVs enough such that taxicab fleets can become EVs. Using fisheye state routing to coordinate rendezvous points, and a simple heuristic where cars exchange charge when there is excess available, we show that we can effectively increase the EV driving distance without needing to recharge. We use real data from the New York City Taxi and Limousine Commission to model taxicab usage behavior and the United States Department of Transportation Omnibus Survey to model commuter vehicle usage. Using this data, we demonstrate that an electric taxicab fleet could function with minimal disturbance (failure <; 5%).

Game theoretic approach to offering participation incentives for electric vehicle-to-vehicle charge sharing

Electric vehicles are not penetrating the market as quickly as expected. This is due to limited driving range, time required to recharge a battery, and lack of charging infrastructure in most metropolitan cities. We propose a charge sharing network in which we use inductive power transfer to wirelessly exchange charge between vehicles. In our network, vehicles that have excess charge to share, can sell charge to vehicles needing charge to reach their destination. In this paper, we describe a game theoretic approach to offering incentives for electric vehicles to participate in the charge sharing network. We utilize Nash Bargaining theory to show that participation in the network can yield profits for the seller driving to their destination and that we can increase the number of cars reaching their destination without needing to stop for recharging.

Audio-based classification of speaker characteristics

The human voice is primarily a carrier of speech, but it also contains non-linguistic features unique to a speaker and indicative of various speaker demographics, e.g. gender, nativity, ethnicity. Such characteristics are helpful cues for audio/video search and retrieval. In this paper, we evaluate the effects of various low-, mid-, and high-level features for effective classification of speaker characteristics. Low-level signal-based features include MFCCs, LPCs, and six spectral features; mid-level statistical features model lowlevel features; and high-level semantic features are based on selected phonemes in addition to mid-level features. Our data set consists of approximately 76.4 hours of annotated audio with 2786 unique speaker segments used for classification. Quantitative evaluation of our method results in accuracy rates up to 98.6% on our test data for male/female classification using mid-level features and a linear kernel support vector machine. We determine that mid- and high-level features are optimal for identification of speaker characteristics.

Case studies of two projects pertaining to information technology and assistive devices

We examine the introduction of engineering design to high school and first-year college students via real community service-learning projects (CSLP). CSLP was chosen because the goal in the course is two-fold: (1) to introduce students to basic engineering principles and (2) to provide them with the opportunity to see the application of such concepts in society. CSLP aims to simulate a design environment in which students with limited technical backgrounds and diverse interests strive to develop viable solutions to actual problems while developing professional skills through client interaction. We provide a detailed discussion on two long-term projects, with approximately 210 student participants, applied both in the high-school and first-year college setting affiliated with the Wildlife Conservation Society and New York City Department of Parks and Recreation, respectively, pertaining to information technology and assistive devices. This approach has been successful in implementation and sustainability. Client satisfaction with the results has lead to the beginning implementation stages. High-school student experiences has led to an increased interest in the pursuit of higher engineering education while first-year college evaluations indicate most felt a renewed interest about their intended major. Future improvements to facilitate sustainability of this program include smoothening the transition process between teams.