Normalized-cut image segmentation
Markov Random Fields
Multi-camera video background subtraction
Patternfinder
Visual Effects in the Flexiview 3D Interactive Video Surveillance System
Random Projection in Machine Learning
In many signal classification domains such as image or video processing, the high dimensionality of the input signals makes the use of standard machine learning algorithms time and computation intensive. Random projection is emerging as a flexible and effective means for dimensionality reduction, particularly when the input signal has a sparse representation in some family of basis functions.
This project examines the effect of the dimensionality of the sparse representation and random projection basis on 2-class separability using several popular machine learning algorithms. In particular, the Discrete Cosine Transform (DCT) basis was chosen as the input space, and SVM, Naive Bayes and Multilayer-Perceptron were used for machine learning. Results showed that the correct classification rate using random projection was comparable to classification in the input domain when the dimensionality of random projection vectors was on the order of the dimensionality of the input domain.
Probabilistic Topic Modeling
Survey of Generative Clustering Models 2008
Survey of Generative Clustering Models 2008 – Download as a PDF or view online for free
Capturing Spatial Context in Images with a Relational Dictionary
Probabilistic generative models such as Probabilistic Latent Semantic Indexing (PLSI), Latent Dirichlet Allocation (LDA), and their variants have been used with some success on the task of object recognition in images. Recent approaches integrate spatial information into the generative process leading to complex hierarchical models. This work takes a simpler approach by encoding the neighborhood of a particular visual word in a relation dictionary. Thus, an image is represented as histogram over this relational dictionary, and standard topic modeling approaches can be applied. We examine three methods of inducing the neighborhood graph and compare them on an image recognition task using the LabelMe benchmark database. Results for the relational dictionary are on-par with a standard non-relational dictionary, but consistently poor, thereby suggesting a flaw in the experimental setup.
COBRAS – A collaborative brainstorming tool
Traditional person-to-person group brainstorming suffers from three major limitations: (1) free riding; (2) productivity blocking; and (3) evaluation apprehension. An ideal electronic brainstorming tool should address these shortcomings while retaining the dynamic, interactive nature of group brainstorming. Moreover, while the nominal goal of brainstorming is to maximize the number of ideas generated, a variety of secondary goals have been proposed as also having value[1]. In particular, it has been suggested that brainstorming helps build organizational memory in terms of design decisions and creates a ‘status auction’. To address these limitations and secondary goals, we have created an online collaborative mind-mapping tool called COBRAS (Collaborative Online Brainstorming by Roman And Saket). Mind-maps are visual representations of ideas and how they are connected to each other[2]. They are usually laid out like a tree which makes it easy to identify relationships and has been proven to help transfer memory into long-term memory and increase recall[2].
References
Robert I. Sutton and Andrew Hargadon. Brainstorming Groups in Context: Effectiveness in a Product Design Firm. Administrative Science Quarterly, Volume 41, Pages 685-718, 1996
Tony Buzan. The Mind Map Book. New York: Penguin, 1991