Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

2018 Cover Image: Thingi10K

2018 Cover Image: Thingi10K “Thingi10K” by Qingnan Zhou and Alec Jacobson.Description: This image is one of the many renditions of the Thingi10K dataset, which includes 10,000 3D models designed for fabrication purposes and were featured by Thingiverse.com from September 16, 2009 to Nov 15, 2015. The dataset was first assembled by Qingnan Zhou (New York University) and Alec Jacobson (Columbia University) back in 2015 for the purpose of stress testing geometry processing algorithms, and they have been curating and analyzing this dataset ever since. Please visit the Thingi10K page (https://ten-thousand-models.appspot.com/) to learn more about these shapes and query for interesting geometric and contextual characteristics associated with them.In this particular rendition, only 6,433 of the 10,000 models were packed randomly into the canvas due to space limitation. The packing algorithm is amazingly simple and algorithmically greedy. Each of the 10,000 models was first rendered separately using the Mitsuba rendering engine. The color of each model was randomly chosen from a small list of hand‐picked colors. Once we have 10,000 renderings, the packing algorithm repeatedly tries to pasted a randomly selected rendering at a random empty spot on the canvas with a randomly chosen scale. If an empty spot is not large enough to avoid http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Computer Graphics Forum Wiley

2018 Cover Image: Thingi10K

Computer Graphics Forum , Volume 37 (1) – Jan 1, 2018

Loading next page...
 
/lp/wiley/2018-cover-image-thingi10k-mCSeGaPk2y

References (1)

Publisher
Wiley
Copyright
© 2018 The Eurographics Association and John Wiley & Sons Ltd.
ISSN
0167-7055
eISSN
1467-8659
DOI
10.1111/cgf.13328
Publisher site
See Article on Publisher Site

Abstract

“Thingi10K” by Qingnan Zhou and Alec Jacobson.Description: This image is one of the many renditions of the Thingi10K dataset, which includes 10,000 3D models designed for fabrication purposes and were featured by Thingiverse.com from September 16, 2009 to Nov 15, 2015. The dataset was first assembled by Qingnan Zhou (New York University) and Alec Jacobson (Columbia University) back in 2015 for the purpose of stress testing geometry processing algorithms, and they have been curating and analyzing this dataset ever since. Please visit the Thingi10K page (https://ten-thousand-models.appspot.com/) to learn more about these shapes and query for interesting geometric and contextual characteristics associated with them.In this particular rendition, only 6,433 of the 10,000 models were packed randomly into the canvas due to space limitation. The packing algorithm is amazingly simple and algorithmically greedy. Each of the 10,000 models was first rendered separately using the Mitsuba rendering engine. The color of each model was randomly chosen from a small list of hand‐picked colors. Once we have 10,000 renderings, the packing algorithm repeatedly tries to pasted a randomly selected rendering at a random empty spot on the canvas with a randomly chosen scale. If an empty spot is not large enough to avoid

Journal

Computer Graphics ForumWiley

Published: Jan 1, 2018

There are no references for this article.