Research Article
Pyramid Algorithm Framework for Real-Time Image Effects in Game Engines
@INPROCEEDINGS{10.1007/978-3-319-55834-9_34, author={Adri\'{a} Arbu\^{e}s Sang\'{y}esa and Andreea-Daniela Ene and Nicolai J\`{u}rgensen and Christian Larsen and Daniel Michelsanti and Martin Kraus}, title={Pyramid Algorithm Framework for Real-Time Image Effects in Game Engines}, proceedings={Interactivity, Game Creation, Design, Learning, and Innovation. 5th International Conference, ArtsIT 2016, and First International Conference, DLI 2016, Esbjerg, Denmark, May 2--3, 2016, Proceedings}, proceedings_a={ARTSIT \& DLI}, year={2017}, month={3}, keywords={Pyramid methods Image effects Depth of field Blur Bloom Game engine Texture lookup Compute shader}, doi={10.1007/978-3-319-55834-9_34} }- Adrià Arbués Sangüesa
Andreea-Daniela Ene
Nicolai Jørgensen
Christian Larsen
Daniel Michelsanti
Martin Kraus
Year: 2017
Pyramid Algorithm Framework for Real-Time Image Effects in Game Engines
ARTSIT & DLI
Springer
DOI: 10.1007/978-3-319-55834-9_34
Abstract
Pyramid methods are useful for certain image processing techniques due to their linear time complexity. Implementing them using compute shaders provides a basis for rendering image effects with reduced impact on performance compared to conventional methods. Although pyramid methods are used in the game industry, they are not easily accessible to all developers because many game engines do not include built-in support. We present a framework for a popular game engine that allows users to take advantage of pyramid methods for developing image effects. In order to evaluate the performance and to demonstrate the framework, a few image effects were implemented. These effects were compared to built-in effects of the same game engine. The results showed that the built-in image effects performed slightly better. The performance of our framework could potentially be improved through optimisation, mainly on the GPU.
