Google has released a new paper that they presented at the recent Computer Vision and Pattern Recognition (CVPR) conference in Miami, FL. Called “Tour the World: building a web-scale landmark recognition engine”, it’s about a new tool they’ve built for image pattern matching to identify images of landmarks.
Our research builds on the vast number of images on the web the ability to search those images and advances in object recognition and clustering techniques. First we generated a list of landmarks relying on two sources 40 million GPS-tagged photos from Picasa and Panoramio and online tour guide webpages. Next we found candidate images for each landmark using these sources and Google Image Search which we then “pruned” using efficient image matching and unsupervised clustering techniques. Finally we developed a highly efficient indexing system for fast image recognition.
They’re adamant that this is not a new google product, just some research, but it’s not hard to imagine this being rolled into something like Picasa in the near future.
via Official Google Blog: A new landmark in computer vision.
At the recent Eurovis conference, Danny Holten and Prof. van Wijk presented a paper entitled “Force-Directed Edge Bundling for Graph Visualization”. They merged the concept of edge bundling with force-directed network graphs, also known as node-link graphs, to show graphs with significantly less clutter and highlight high-level edge patterns. The picture above models the edges as flexible springs attracted to each other.
Get the paper here.
Force-Directed Edge Bundling for Graph Visualization – information aesthetics.
I just found this paper by Yuko Uematsu and Hideo Saito of the Graduate School of Science and Technology in Japan. Titled “Visual Enhancement for Sports Entertainment by Vision-Based Augmented Reality”, it discusses ways that Augmented Reality can aid in simple sports games. Two specific examples include replays of classic Baseball games, and an interactive bowling game.
AR Baseball Presentation System is an observation system of a virtual baseball game. Users place a real baseball field model on the tabletop and input a baseball game history (scorebook) that they want to watch into the system. Then, they can watch the game by replaying with virtual baseball players on the field model in front of them. On the field model, 2D markers are placed for registration of the virtual players. Therefore, the users can watch the game from their favorite viewpoints around the field.
The entire paper is available at Hindawi publishing as a PDF.
via Visual Enhancement for Sports Entertainment by Vision-Based Augmented Reality.
While not traditionally what you’ld call Information Visualization, the above graphic is a powerful visualization of William Meehan’s dissertation, with portions highlighted that are copied verbatim from Carl Boening’s dissertation. The two universities involved (Jacksonville State University and the University of Alabama) claim it’s not plagiarism.
Some enterprising readers (faculty? student-journalists?) have gone through the dissertations of Carl Boening and William Meehan, highlighting every passage in Meehan’s that can be found, word for word, in Boening’s. Neither the University of Alabama (which granted Boening and Meehan their doctorates) nor Jacksonville State University, where Meehan is president, has chosen to take up the obvious questions about plagiarism that Meehan’s dissertation presents. As another recent story suggests, plagiarism seems to be governed by a sliding scale, with consequences lessening as the wrongdoer’s status rises.
via Orange Crate Art: What plagiarism looks like.
I have been thinking about making a list of some of the most seminal information visualization papers. These are papers that have made an impact and can be widely seen in the media (print/web) or are being adopted in visualization software/systems such as VTK, Prefuse, Many Eyes and so on. I may have missed out on a few papers, so please feel free to add any that you think are ‘must-reads’ for an infovis researcher.
Click through for the list…
SIGGRAPH has announced some of the papers that will appear in the Technical Papers program, and it’s a pretty impressive collection.
Papers cover core topics of computer graphics, such as modeling, animation, rendering, imaging, and human-computer interaction, and also explore related fields of audio, robotics, visualization, and perception. Presenters are from all around the globe – from the Czech Republic to Japan.
The list so far:
- UC Berkeley & UNC Chapel Hill – Interactive Simulation of Surgical Needle Insertion and Steering
- MIT – Bokode: Imperceptible Visual Tags for Camera-based Interaction from a Distance
- NYU – Dark Flash Photography
- MIT, Adobe, and UW – Real-Time Hand-Tracking with a Color Glove
- Cornell – Harmonic Fluids
- ILM – Directable, High-Resolution Simulation of Fire on the GPU
Looks like a good group, especially the ILM paper.
via SIGGRAPH 2009 Technical Papers Focus on Technology and Advanced Techniques.