Submissions to the Technical Papers program are received from around the globe and feature high quality, never-before-seen scholarly work. Submitters are held to extremely high standards in order to qualify.
“The quality of this year’s Technical Papers content was superb,” said Adam Finkelstein, SIGGRAPH 2014 Technical Papers Chair. “Not only did we receive outstanding work in the traditional areas of modeling, rendering, animation, and imaging, but we also noticed two emerging fields in design/fabrication and learning.”
SIGGRAPH accepted 127 technical papers (out of 550 submissions) for this year’s showcase, an acceptance rate of 25 percent (one percent higher than 2013). As is tradition, the papers were chosen by a committee made up of academia and industry experts.
The 2014 Technical Papers program also includes presentations for 46 papers published this year in the journal ACM Transactions on Graphics (TOG).
Highlights From the SIGGRAPH 2014 Technical Papers Program inlcude:
Pixie Dust: Graphics Generated by Levitate and Animated Objects in Computational Acoustic-Potential Field
Authors: Yoichi Ochiai, Jun Rekimoto, The University of Tokyo; Takayuki Hoshi, Nagoya Institute of Technology
To generate graphics using levitated small objects, this method expands the acoustic manipulation method by changing the distribution of the acoustic-potential field. The approach makes available many expressions (for example, expression by materials and nondigital appearance), and the expressions are levitated in mid-air.
Floating Scale Surface Reconstruction
Authors: Simon Fuhrmann, Michael Goesele, Technische Universität Darmstadt
Introducing a novel, virtually parameter-free method for surface reconstruction from oriented, scale-enabled point samples. The approach constructs an implicit function as the sum of compactly supported basis functions. The final surface is extracted as the zero-level set of the implicit function.
Authors: Markus Lipp, Pascal Müller, Esri; Peter Wonka, King Abdullah University of Science and Technology
Introducing the novel 3D modeling tool PushPull++, which enables rapid modeling and adjustment of arbitrary meshes with slanted surfaces. PushPull++ reduces the required clicks for common modeling tasks up to an order of magnitude compared to leading commercial tools.
Intrinsic Images in the Wild
Authors: Sean Bell, Kavita Bala, Noah Snavely, Cornell University
Introducing a large-scale, public dataset for intrinsic image decompositions of indoor scenes (Intrinsic Images in the Wild) with crowdsourced annotations of pairwise comparisons of material properties. The approach develops a dense CRF-based algorithm for intrinsic image decomposition of these images that outperforms state-of-the-art algorithms.
Learning Bicycle Stunts
Authors: Jie Tan, Yuting Gu, Karen Liu, Greg Turk, Georgia Institute of Technology
A general approach to simulation and control of a human character riding a bicycle. The rider not only learns to steer and balance in normal riding situations, but also learns to perform a wide variety of stunts, including wheelie, endo, bunny hop, front-wheel pivot, and back hop.