---
title: "Learning Consumer Photo Categories for Semantic Retrieval"
authors:
  - "Joo-Hwee Lim"
  - "Jesse S. Jin"
doi: ""
url: "https://ijcai.org/proceedings/2003/219"
pdf: "https://ijcai.org/proceedings/2003/0219.pdf"
venue: "Proceedings of the Eighteenth International Joint Conference on Artificial Intelligence"
year: 2003
pages: "1413-1413"
track: ""
keywords:
---

# Learning Consumer Photo Categories for Semantic Retrieval

Joo-Hwee Lim, Jesse S. Jin

*Proceedings of the Eighteenth International Joint Conference on Artificial Intelligence*, pages 1413-1413.

## Abstract

In this paper, we develop a computational learning framework to build a hierarchy of 11 cons um er photo categories for semantic retrieval. Two levels of visual semantics are learned for image content and image category statist ic ally. We evaluate the average precisions at top retrieved photos on 2400 heterogeneous consumer photos with very good result.