Even in the absence of any explicit semantic annotation, vast collections of audio recordings provide valuable information for learning the categorical structure of sounds. We consider several classagnostic semantic constraints that apply to unlabeled nonspeech audio: (i) noise and translations in time do not change the underlying sound category, (ii) a mixture of two sound events inherits the ...

Imitation is a powerful mechanism whereby knowledge may be transferred between agents (both biological and artificial). Key problems on the topic of imitation have emerged in various areas close to artificial intelligence, including the cognitive and social sciences, animal behavior, robotics, human–computer interaction, embodied intelligence, software engineering, programming by example and ma...

We present an algorithm that approximates the output of an arbitrary video processing algorithm based on a pair of input and output exemplars. Our algorithm relies on learning the mapping between the input and output exemplars to model the processing that has taken place. We approximate the processing by observing that pixel neighborhoods similar in appearance and motion to those in the exempla...

Adaptive Programming Environment (APE), a software assistant embedded into the VisualWorks Smalltalk interactive programming environment, watches what the user is doing, draws on machine learning to learn the user’s habits, and afterward offers to complete repetitive tasks on his or her behalf. The goal of the APE project was threefold: (1) to design an assistant able to automate repetitive tas...

Animated characters bring the illusion of life to feature films, television shows, video games, and educational simulations. However, it is difficult for artists to define natural and expressive movement. This challenge is compounded by the fact that people are intrinsically sensitive to subtle inaccuracies in human motion. To address this problem, this thesis presents techniques that adjust ex...

In this paper, we extend the conventional concept of a database as a set of discrete relations to include a set of piecewise continuous functions. We extend the features of Query-by-Example to operations on this piecewise continuous data. Further, we include the concept of iteration in the language, which enhances its capabilities to that of a general programming language. These extensions are ...

Processing queries for databases containing multimedia information is a hard question. In this paper we describe a knowledge base approach to effectively resolve query by examples in image database. We present a prototype system that is at the moment under developing in our Labs. We consider wavelet descriptors for modeling color, shape and texture features, and we propose a knowledge base for ...

This paper summarizes our work for MediaEval 2013 Spoken Web Search task evaluations. The task was Query-by-Example (search of spoken queries within spoken data). We submitted a system composed of 26 subsystems, of which 13 are based on Acoustic Keyword Spotting and 13 on Dynamic Time Warping. All of them use threestate phoneme posteriors as input features. Our main contribution was m-norm norm...