Abstract
We propose a facial video frame interpolation algorithm, which combines symmetric and asymmetric motions to complement each other. We generate multi-scale conual features and intermediate frame candidates using both kinds of motion vectors. Then, through a frame synthesis block, we refine those intermediate frame candidates, by exploiting the conual features, to produce the final output. Extensive experimental results demonstrate that the proposed algorithm provides state-of-the-art performance by combining both symmetric and asymmetric motion information effectively.
| Original language | English |
|---|---|
| Title of host publication | 2021 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2021 - Proceedings |
| Publisher | Institute of Electrical and Electronics Engineers Inc. |
| Pages | 164-169 |
| Number of pages | 6 |
| ISBN (Electronic) | 9789881476890 |
| Publication status | Published - 2021 |
| Event | 2021 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2021 - Tokyo, Japan Duration: 2021 Dec 14 → 2021 Dec 17 |
Publication series
| Name | 2021 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2021 - Proceedings |
|---|
Conference
| Conference | 2021 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2021 |
|---|---|
| Country/Territory | Japan |
| City | Tokyo |
| Period | 21/12/14 → 21/12/17 |
Bibliographical note
Funding Information:This work was supported partly by Samsung Electronics Co., Ltd. (No. IO201214-08156-01) and partly by the National Research Foundation of Korea (NRF) grants funded by the Korea government (MSIT) (No. NRF-2018R1A2B3003896 and No. NRF-2021R1A4A1031864).
Publisher Copyright:
© 2021 APSIPA.
ASJC Scopus subject areas
- Artificial Intelligence
- Computer Vision and Pattern Recognition
- Signal Processing
- Instrumentation