A tree-based approach to integrated action segmentation, localization and recognition is proposed. An action is represented as a sequence of joint hog-flow descriptors extracted independently from each frame. During training, a set of action prototypes is first learned based on a k-means clustering, and then a binary tree model is constructed from the set of action prototypes based on hierarchi...

Given a set S of n points in lRd, we show, for fixed d, how to construct in O(n log n) time a data structure we call the Balanced Aspect Ratio (BAR) tree. A BAR tree is a binary space partition tree on S that has O(logn) depth and in which every region is convex and “fat” (that is, has a bounded aspect ratio). While previous hierarchical data structures, such as k-d trees, quadtrees, octrees, f...

When initially reported [1], BTPC was compared with JPEG, GIF and research coders of that time, SPIHT and CALIC. It outperformed JPEG and GIF (except for limited-palette colour graphics) but was inferior to SPIHT and CALIC in their respective areas of strength. Tested with an assortment of images of different modalities, BTPC was the most consistent overall performer, other coders faring poorly...

The problem this paper focuses on is the classical problem of unsupervised clustering of a data-set. In particular, the bisecting divisive clustering approach is here considered. This approach consists in recursively splitting a cluster into two sub-clusters, starting from the main data-set. This is one of the more basic and common problems in fields like pattern analysis, data mining, document...

We demonstrate efficient approximate inference for the Dirichlet Diffusion Tree (Neal, 2003), a Bayesian nonparametric prior over tree structures. Although DDTs provide a powerful and elegant approach for modeling hierarchies they haven’t seen much use to date. One problem is the computational cost of MCMC inference. We provide the first deterministic approximate inference methods for DDT model...

For fixed k ≥ 2 and fixed data alphabet of cardinality m, the hierarchical type class of a data string of length n = k for some j ≥ 1 is formed by permuting the string in all possible ways under permutations arising from the isomorphisms of the unique finite rooted tree of depth j which has n leaves and k children for each non-leaf vertex. Suppose the data strings in a hierarchical type class a...

We propose solutions to enhance the Inside-Outside Recursive Neural Network (IORNN) reranker of Le and Zuidema (2014). Replacing the original softmax function with a hierarchical softmax using a binary tree constructed by combining output of the Brown clustering algorithm and frequency-based Huffman codes, we significantly reduce the reranker’s computational complexity. In addition, enriching c...

Discovering hierarchical regularities in data is a key problem in interacting with large datasets, modeling cognition, and encoding knowledge. A previous Bayesian solution—Kingman’s coalescent—provides a probabilistic model for data represented as a binary tree. Unfortunately, this is inappropriate for data better described by bushier trees. We generalize an existing belief propagation framewor...