FaceSyncNet : a deep learning-based approach for non-linear synchronization of facial performance videos = 얼굴 표정 연기 비디오의 비선형 동기화를 위한 심층 학습 기반 접근법a deep learning-based approach for non-linear synchronization of facial performance videos

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 52
  • Download : 0
Given a pair of facial performance videos, we present a deep learning-based approach that can automatically return a synchronized version of these videos. Traditional methods require precise facial landmark tracking and/or clean audio, and thus are sensitive to tracking inaccuracies and audio noise. To alleviate these issues, our approach leverages large-scale video datasets along with their associated audio tracks and trains a deep learning network to learn the audio descriptors of video frames. We then use these descriptors to compute the similarity between video frames in a cost matrix and compute a low-cost non-linear synchronization path. Both quantitative and qualitative evaluations have shown that our approach outperforms existing state-of-the-art methods.
Advisors
Lee, Sung-Heeresearcher이성희researcher
Description
한국과학기술원 :문화기술대학원,
Publisher
한국과학기술원
Issue Date
2020
Identifier
325007
Language
eng
Description

학위논문(석사) - 한국과학기술원 : 문화기술대학원, 2020.2,[iii, 23 p. :]

Keywords

Video synchronization▼aDeep learning▼aFacial performance video▼aAudio visual information; 비디오 동기화▼a딥러닝▼a얼굴 표정 연기 비디오▼a음성-시각 정보

URI
http://hdl.handle.net/10203/284019
Link
http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=910812&flag=dissertation
Appears in Collection
GCT-Theses_Master(석사논문)
Files in This Item
There are no files associated with this item.

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0