PhotoMesa A Zoomable Image Browser Using Quantum Treemap and Bubblemap

1
PhotoMesaA Zoomable Image Browser Using Quantum
Treemap and Bubblemap

• Presented by Khanh-Dung (KD) Nguyen
• COSC 686 Information Visualization
• Instructor Dr. Harry Hochheiser
• 02/27/2007

2
Introduction

• Information retrieval systems for
  multimediaimages
• 2001 queries for items of interest
• Need good interface, visualization interaction
  techniques
• Importance
• Need for browsing of search results, (no grids of
  thumbnails, scroll bars)
• Pleasure spontaneity. (no custom albums)

3
The Heart of PhotoMesa

• I wanted to concentrate on the images and more
  importantly, as I was looking at the photos with
  my daughter, it was crucial that she be an active
  part of the interaction, and not just a passive
  bystander.
• Benjamin B. Bederson

4
Related Works

• PhotoFinder - polished traditional approach
• Document Lens 2D Fisheye view - single
  view at a time
• AutoAlbum automatically cluster photos -
  scrolling
• Personal Digital Historian co-present
  use - multiple images interaction
• ZoomBrowser EX ZUIs
• Pad - one sized directories
  ? unreadable

5
Design Goals

• Simple to use
• Work well for family-use settings
• Support collections of photos, efficient use
  of screen space.

6
New Algorithms

• Quantum treemap
• Treemap produces proportional areas of 2D space
  according to requested list
• Quantum indivisible contents
• Bubblemap
• Generates non-rectangular groups (of images)

7
Quantum Treemap

• Stable aspect ratio
• Grow horizontally / vertically
• ? best matches aspect ratio
• Stopping conditions
• Regular
• Quad 2x2 rectangles
• Snake sequential lay-out
• Pivot selection
• try all 3 lay-outs (middle, largest or smallest
  element ? closest to equal size

8
Quantum vs. Ordered Treemap
9
Bubblemap

• Minimum unused space
• Good with pre-clustered high-dimensional set of
  relationships
• Fast O(n) run time
• ? Could be hard to parse visually

10
Future Work

• Integration with PhotoFinder other metadata
• Deployment with different modalities (stand-alone
  CDs, integration with the Web)
• Detailed user study for effectiveness
  efficiency.

11
References

• Balabanovic, M., Chu, L. L., Wolff, G. J.
  (2000). Storytelling with Digital Photographs. In
  proceedings of Human Factors in Computing Systems
  (CHI 2000) ACM Press, pp.564-571.
• Kang, H., Schneiderman, N. (2000).
  Visualization Methods for Personal Photo
  Collections Browsing and Searching in the
  PhotoFinder. In Proceedings of IRRR International
  Conference on Multimedia and Expo (ICME2000) New
  York IEEE, pp. 1539-1342.
• Platt, J. (2000). AutoAlbum Clustering Digital
  Photographs Using Probabalistic Model Merging. In
  Proceedings of IEEE Workshop on Content-based
  Access of Image and Video Libraries (CBAIVL-2000)
  IEEE Press
• Robertson, G. G., Mackinlay, J. D. (1993). The
  Document Lens. In Proceedings of User Interface
  and Software Technology (UIST 93) ACM Press,
  pp.101-108.
• Shen, C. Moghaddam, B., Lesh, N., Beardsley, P.
  (2001). Personal Digital Historian User
  Interface Design. In Proceedings of Extended
  Abstracts of Human Factors in Computing Systems
  (CHI 2001) ACM Press

12
Discussion

• Questions?
• Thoughts?
• Comments?
• Thanks!