Screenshot of the website wikiview.net

wikiview.net

Wikiview revolutionizes image search by organizing Wikimedia Commons images on an interactive 2D map based on visual similarity, offering a unique and intuitive browsing experience.
Visit wikiview.net
Wikiview enables users to explore the extensive Wikimedia Commons image collection through a seamless interface combining color, text, and image search. The hierarchical data structure arranges visually similar photos side by side, offering an intuitive and in-depth browsing experience.

wikiview.net, an award-winning technology, has been superseded by navigu.net. All data and functionality from wikiview.net are now fully integrated into navigu.net, and as a result, wikiview.net has been discontinued.

Key Features

Visual Image Map: Wikiview presents search results on a visually organized map where similar images are grouped together. This layout allows users to easily discover related photos without needing to scroll through long lists.

Multi-Modal Search: Users can search by text, color, or by uploading an image to find visually similar results. This flexibility enables more precise searches and helps in finding exactly the right image quickly.

Dynamic Refinement: As users explore the map, they can add additional search terms or adjust filters to refine results on the fly. This dynamic approach ensures that the search process remains intuitive and responsive to specific needs.

How to Use Wikiview?

  1. Start a Search: Enter a keyword, select a color, or upload an image to begin your search.
  2. Explore the Map: Navigate through the visually organized 2D map where similar images are grouped together.
  3. Refine Your Results: Add additional search terms or adjust filters to narrow down your results as needed.
  4. View and Download: Click on any image to see detailed information, and download it directly if desired.
  5. Discover More: Use the sidebar to explore similar images or follow links to their Wikimedia Commons pages for further exploration.

How Does Wikiview Work?

Wikiview organizes its image collection as a graph, where each image is represented as a vertex connected to similar images. During a search, the system identifies the best matching image and follows its edges to find related images. These related images are displayed on a visually sorted map, which users can explore by dragging and zooming. As users navigate the map, new images are dynamically retrieved and seamlessly integrated into the empty areas. This approach ensures a smooth and continuous exploration experience, allowing users to discover images efficiently.

Watch the video on YouTube

Demonstration of the functionality of wikiview.net

Publications

List of relevant publications

Real-Time Visual Navigation in Huge Image Sets Using Similarity Graphs

Kai Uwe Barthel, Nico Hezel, Konstantin Schall, Klaus Jung
27th ACM International Conference on Multimedia (MM '19)

We demonstrate a system for real-time visual exploration of millions of images using a standard web browser, addressing the challenge of inspecting large collections. Our approach leverages graph-based image navigation and a hierarchical similarity graph to present images in a dynamically updated, zoomable 2D map.

Visual Navigation of Large Image Graphs

Nico Hezel, Kai Uwe Barthel, Konstantin Schall, Klaus Jung
2019 IEEE 21st International Workshop on Multimedia Signal Processing (MMSP '19)

We present a graph-based system for visually exploring and navigating millions of images in a web browser. This demo retrieves subsets of images from a similarity graph and displays them as a zoomable, draggable 2D map.

Visually Exploring Millions of Images using Image Maps and Graphs

Kai Uwe Barthel, Nico Hezel
Big Data Analytics for Large‐Scale Multimedia Search, Chapter 11, John Wiley & Sons, Ltd (2019)

This work introduces various image sorting algorithms and a new measure for evaluating 2D image arrangements. It presents an enhanced self-sorting maps (SSM) algorithm for efficiently sorting millions of images, discusses a graph-based approach for browsing large collections, and explains how high-quality image features are generated using convolutional neural networks.