| Title |
Micro Expression Generation using Diffusion Model and Optical Flow |
| Authors |
김찬호(Chanho Kim) ; 박인규(In Kyu Park) |
| DOI |
https://doi.org/10.5573/ieie.2023.60.10.43 |
| Keywords |
Micro expression; Deep-learning; Generative model |
| Abstract |
Micro-expressions are small movements of the face that can be difficult to manipulate as desired, and can reveal true emotions or states that one might wish to hide. However, related research is difficult due to lack of data. In this paper, we propose a micro-expression synthesis technique as a method to improve this. The proposed method uses a diffusion model and optical flow to generate micro-expressions for input face video, given micro-expression videos and new face videos as inputs. Through this, it is possible to improve the data shortage of micro-expressions, We prove the effect of improving the performance of micro-expression recognition by including the data generated through the proposed method in the training data. |