Note: Your custom schedule will not be saved unless you create a new account or login to an existing account.
| Paper: | IMDSP-P7.12 |
| Session: | Indexing and Retrieval |
| Time: | Wednesday, May 19, 15:30 - 17:30 |
| Presentation: |
Poster |
| Topic: |
Image and Multidimensional Signal Processing: Image and Video Indexing and Retrieval |
| Title: |
INTERACTIVE IMAGE RETRIEVAL BY QUERY FUSION |
| Authors: |
Azadeh Kushki; University of Toronto | | |
| | Panagiotis Androutsos; University of Toronto | | |
| | Anastasios Venetsanopoulos; University of Toronto | | |
| Abstract: |
The use of low-level visual features such as colour and shape incontent-based image retrieval system leads to several ambiguities.Specifically, due to the many-to-many mapping between thelow-level feature space and high-level user concepts, a conceptualuser query may not be modelled as a single point in the featurespace. Furthermore, conceptually similar images may not be fallclose to each other in the low-level feature space. This workaddresses these issues by proposing an interactive technique,Query Feedback. This method employs user input to represent ahigh-level conceptual query as various points in the low-levelspace. Thus, similarity to a given conceptual query is obtained asthe fusion of several low-level representations in the featurespace. |
| |
| Back | |