Structure Inference Machines: Recurrent Neural Networks for Analyzing Relations in Group Activity Recognition release_exnlbwkkgrh3pj2eyg5i7jxu2i

by Zhiwei Deng, Arash Vahdat, Hexiang Hu, Greg Mori

Released as a article .

2015  

Abstract

Rich semantic relations are important in a variety of visual recognition problems. As a concrete example, group activity recognition involves the interactions and relative spatial relations of a set of people in a scene. State of the art recognition methods center on deep learning approaches for training highly effective, complex classifiers for interpreting images. However, bridging the relatively low-level concepts output by these methods to interpret higher-level compositional scenes remains a challenge. Graphical models are a standard tool for this task. In this paper, we propose a method to integrate graphical models and deep neural networks into a joint framework. Instead of using a traditional inference method, we use a sequential inference modeled by a recurrent neural network. Beyond this, the appropriate structure for inference can be learned by imposing gates on edges between nodes. Empirical results on group activity recognition demonstrate the potential of this model to handle highly structured learning tasks.
In text/plain format

Archived Files and Locations

application/pdf  710.5 kB
file_gtp636e3xnhk5bc7bcwfvtwq5q
arxiv.org (repository)
web.archive.org (webarchive)
Read Archived PDF
Preserved and Accessible
Type  article
Stage   submitted
Date   2015-11-13
Version   v1
Language   en ?
arXiv  1511.04196v1
Work Entity
access all versions, variants, and formats of this works (eg, pre-prints)
Catalog Record
Revision: 31b13b96-1c64-46a3-91d4-34bfa7e055c5
API URL: JSON