Beyond Actions: Discriminative Models for Contextual Group Activities
نویسندگان
چکیده
Inference: We approximately solve the inference problem by iterating the following two steps: 1. Holding Gy fixed, optimize hy (solved by Loopy BP): hy = argmax h’ wΨ(x,h’, y;Gy) 2. Holding hy fixed, optimize Gy (solved by integer linear program (ILP)): Gy = argmax G ′ wΨ(x,hy, y;G ) We define a variable z, zjk = 1 indicates that the edge (j, k) is included in the graph, and 0 otherwise. we enforce graph sparsity by setting a threshold d on the maximum degree of any vertex in the graph. Then step 2 can be formulated as an ILP: max z ∑
منابع مشابه
During the past Decade, Computer Vision Research Has Focused on Constructing beyond Nouns and Verbs
During the past decade, computer vision research has focused on constructing image based appearance models of objects and action classes using large databases of examples (positive and negative) and machine learning to construct models. Visual inference however involves not only detecting and recognizing objects and actions but also extracting rich relationships between objects and actions to f...
متن کاملTitle of dissertation : BEYOND NOUNS AND VERBS
Title of dissertation: BEYOND NOUNS AND VERBS Abhinav Gupta, Doctor of Philosophy, 2009 Dissertation directed by: Professor Larry S. Davis Department of Computer Science During the past decade, computer vision research has focused on constructing image based appearance models of objects and action classes using large databases of examples (positive and negative) and machine learning to construc...
متن کاملActivity Group Localization by Modeling the Relations among Participants
Beyond recognizing the actions of individuals, activity group localization aims to determine “who participates in each group” and “what activity the group performs”. In this paper, we propose a latent graphical model to group participants while inferring each group’s activity by exploring the relations among them, thus simultaneously addressing the problems of group localization and activity re...
متن کاملSupervised Topic Models for Video Activity Recognition
Topic models successfully capture latent structure useful for unsupervised analysis of bag-of-words data. Applying these models to domains such as video activity recognition requires two critical extensions: (1) incorporating supervised information (activity labels) to recover topic structure with greater discriminative power and (2) moving beyond the bag-of-words assumption to model temporal d...
متن کاملA Discriminative Lexicon Model for Complex Morphology
This paper describes successful applications of discriminative lexicon models to the statistical machine translation (SMT) systems into morphologically complex languages. We extend the previous work on discriminatively trained lexicon models to include more contextual information in making lexical selection decisions by building a single global log-linear model of translation selection. In offl...
متن کامل