Loading...
Please wait, while we are loading the content...
Similar Documents
Nechyba, “Interpretation of complex scenes using generative dynamic-structured models,” in (to appear (2004)
| Content Provider | CiteSeerX |
|---|---|
| Author | Todorovic, Sinisa Nechyba, Michael C. |
| Description | We propose a generative modeling framework – namely, Dynamic Tree Structured Belief Networks (DTSBNs) and a novel Structured Variational Approximation (SVA) inference algorithm for DTSBNs – as a viable solution to object recognition in images with partially occluded object appearances. We show that it is possible to assign physical meaning to DTSBN structures, such that root nodes model whole objects, while parent-child connections encode component-subcomponent relationships. Therefore, within the DTSBN framework, the treatment and recognition of object parts requires no additional training, but merely a particular interpretation of the tree/subtree structure. As such, DTSBNs naturally allow for multi-stage object recognition, in which initial recognition of object parts induces recognition of objects as a whole. As our reported experiments show, this explicit, multi-stage treatment of occlusion outperforms more traditional objectrecognition approaches, which typically fail to account for occlusion in any principled or unified manner. 1. |
| File Format | |
| Language | English |
| Publisher Date | 2004-01-01 |
| Publisher Institution | Proc. IEEE CVPR 2004, Workshop on GenerativeModel Based Vision (GMBV |
| Access Restriction | Open |
| Subject Keyword | Multi-stage Treatment Physical Meaning Occlusion Outperforms Component-subcomponent Relationship Whole Object Parent-child Connection Object Part Induces Recognition Complex Scene Dtsbn Framework Tree Subtree Structure Additional Training Object Appearance Dtsbn Structure Multi-stage Object Recognition Dynamic Tree Structured Belief Network Inference Algorithm Generative Dynamic-structured Model Particular Interpretation Initial Recognition Traditional Objectrecognition Approach Viable Solution Object Part Unified Manner Novel Structured Variational Approximation Root Node Experiment Show Generative Modeling Framework |
| Content Type | Text |
| Resource Type | Article |
