---
title: "SIGMA: A Framework for Image Understanding - Integration of Bottom-Up and Top-Down Analysis"
authors:
  - "Takashi Matsuyama"
  - "Vincent Shang-Shouq Hwang"
doi: ""
url: "https://ijcai.org/proceedings/1985/183"
pdf: "https://ijcai.org/proceedings/1985/0183.pdf"
venue: "Proceedings of the Ninth International Joint Conference on Artificial Intelligence"
year: 1985
pages: "908-914"
track: "Perception"
keywords:
---

# SIGMA: A Framework for Image Understanding - Integration of Bottom-Up and Top-Down Analysis

Takashi Matsuyama, Vincent Shang-Shouq Hwang

*Proceedings of the Ninth International Joint Conference on Artificial Intelligence*, pages 908-914.

## Abstract

The framework and control structure of an Image understanding system SIGMA are presented. SIGMA consists of three experts: Geometrie Reasoning Expert (GRE) for spatial reasoning, Model Selection Expert (MSE) for appearance model selection, and Low Level Vision Expert (LLVE) for knowledge-based picture processing. This paper mainly describes the control mechanism f01· the spatial reasoning by GRE, where bottom-up and top-down analyses are integrated into a unified reasoning process.