Search
📃

Two-step Temporal Interpolation Network Using Forward Advection for Efficient Smoke Simulation

•
Citation
Oh, Young Jin, and In‐Kwon Lee. "Two‐step Temporal Interpolation Network Using Forward Advection for Efficient Smoke Simulation." Computer Graphics Forum. Vol. 40. No. 2. 2021.
•
Abstract
In this paper, we propose a two-step temporal interpolation network using forward-advection for efficient smoke simulation generation. Since a low frame rate smoke simulation calculated with a large time-step is converted into a high frame rate smoke simulation by inference of temporal interpolation networks, the proposed method could efficiently generate a smoke simulation result with a high frame rate with low computational costs. The proposed method's first step is to perform optical flow-based temporal interpolation using Deep Neural Networks(DNN) for two given smoke animation frames. In the next second step, we calculate temporary smoke frames with forward-advection, a physical calculation with a low computational cost. We then interpolate between the results of the forward-advection and the results of the first step to generate more accurate and enhanced interpolation results. We performed quantitative analyses of the results generated by the proposed method and previous temporal interpolation methods. Furthermore, we compared the proposed method's performance with previous methods using DNN to generate smoke simulation. As a result of the experiment, we found that the results generated by the proposed method are more accurate and closer to the ground truth smoke simulation than those generated by the previous temporal interpolation methods. We also confirmed that the proposed method generates smoke simulation results more efficiently with lower computation costs than previous smoke simulation methods using DNN. (Journal IF = 2.078 (2020), Ranking = 45.37% (Q2), Category = COMPUTER SCIENCE - SOFTWARE ENGINEERING, SCIE)
•
Videos