Matches in SemOpenAlex for { <https://semopenalex.org/work/W4288329168> ?p ?o ?g. }
Showing items 1 to 47 of
47
with 100 items per page.
- W4288329168 abstract "Learning to synthesize high frame rate videos via interpolation requires large quantities of high frame rate training videos, which, however, are scarce, especially at high resolutions. Here, we propose unsupervised techniques to synthesize high frame rate videos directly from low frame rate videos using cycle consistency. For a triplet of consecutive frames, we optimize models to minimize the discrepancy between the center frame and its cycle reconstruction, obtained by interpolating back from interpolated intermediate frames. This simple unsupervised constraint alone achieves results comparable with supervision using the ground truth intermediate frames. We further introduce a pseudo supervised loss term that enforces the interpolated frames to be consistent with predictions of a pre-trained interpolation model. The pseudo supervised loss term, used together with cycle consistency, can effectively adapt a pre-trained model to a new target domain. With no additional data and in a completely unsupervised fashion, our techniques significantly improve pre-trained models on new target domains, increasing PSNR values from 32.84dB to 33.05dB on the Slowflow and from 31.82dB to 32.53dB on the Sintel evaluation datasets." @default.
- W4288329168 created "2022-07-28" @default.
- W4288329168 creator A5016107863 @default.
- W4288329168 creator A5025713692 @default.
- W4288329168 creator A5033212845 @default.
- W4288329168 creator A5041525280 @default.
- W4288329168 creator A5044231615 @default.
- W4288329168 creator A5056503617 @default.
- W4288329168 creator A5066242985 @default.
- W4288329168 creator A5072436307 @default.
- W4288329168 creator A5089901400 @default.
- W4288329168 date "2019-06-13" @default.
- W4288329168 modified "2023-09-28" @default.
- W4288329168 title "Unsupervised Video Interpolation Using Cycle Consistency" @default.
- W4288329168 doi "https://doi.org/10.48550/arxiv.1906.05928" @default.
- W4288329168 hasPublicationYear "2019" @default.
- W4288329168 type Work @default.
- W4288329168 citedByCount "0" @default.
- W4288329168 crossrefType "posted-content" @default.
- W4288329168 hasAuthorship W4288329168A5016107863 @default.
- W4288329168 hasAuthorship W4288329168A5025713692 @default.
- W4288329168 hasAuthorship W4288329168A5033212845 @default.
- W4288329168 hasAuthorship W4288329168A5041525280 @default.
- W4288329168 hasAuthorship W4288329168A5044231615 @default.
- W4288329168 hasAuthorship W4288329168A5056503617 @default.
- W4288329168 hasAuthorship W4288329168A5066242985 @default.
- W4288329168 hasAuthorship W4288329168A5072436307 @default.
- W4288329168 hasAuthorship W4288329168A5089901400 @default.
- W4288329168 hasBestOaLocation W42883291681 @default.
- W4288329168 hasConcept C115961682 @default.
- W4288329168 hasConcept C137800194 @default.
- W4288329168 hasConcept C154945302 @default.
- W4288329168 hasConcept C2776436953 @default.
- W4288329168 hasConcept C31972630 @default.
- W4288329168 hasConcept C41008148 @default.
- W4288329168 hasConceptScore W4288329168C115961682 @default.
- W4288329168 hasConceptScore W4288329168C137800194 @default.
- W4288329168 hasConceptScore W4288329168C154945302 @default.
- W4288329168 hasConceptScore W4288329168C2776436953 @default.
- W4288329168 hasConceptScore W4288329168C31972630 @default.
- W4288329168 hasConceptScore W4288329168C41008148 @default.
- W4288329168 hasLocation W42883291681 @default.
- W4288329168 hasOpenAccess W4288329168 @default.
- W4288329168 hasPrimaryLocation W42883291681 @default.
- W4288329168 isParatext "false" @default.
- W4288329168 isRetracted "false" @default.
- W4288329168 workType "article" @default.