FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2892008670> ?p ?o ?g. }

• W2892008670 abstract "Compositing is one of the most important editing operations for images and videos. The process of improving the realism of composite results is often called harmonization. Previous approaches for harmonization mainly focus on images. In this work, we take one step further to attack the problem of video harmonization. Specifically, we train a convolutional neural network in an adversarial way, exploiting a pixel-wise disharmony discriminator to achieve more realistic harmonized results and introducing a temporal loss to increase temporal consistency between consecutive harmonized frames. Thanks to the pixel-wise disharmony discriminator, we are also able to relieve the need of input foreground masks. Since existing video datasets which have ground-truth foreground masks and optical flows are not sufficiently large, we propose a simple yet efficient method to build up a synthetic dataset supporting supervised training of the proposed adversarial network. Experiments show that training on our synthetic dataset generalizes well to the real-world composite dataset. Also, our method successfully incorporates temporal consistency during training and achieves more harmonious results than previous methods." @default.
• W2892008670 created "2018-09-27" @default.
• W2892008670 creator A5027959479 @default.
• W2892008670 creator A5029588621 @default.
• W2892008670 creator A5037233582 @default.
• W2892008670 creator A5052835435 @default.
• W2892008670 creator A5071037763 @default.
• W2892008670 date "2018-09-05" @default.
• W2892008670 modified "2023-09-24" @default.
• W2892008670 title "Temporally Coherent Video Harmonization Using Adversarial Networks" @default.
• W2892008670 cites W1535390478 @default.
• W2892008670 cites W1861492603 @default.
• W2892008670 cites W1901129140 @default.
• W2892008670 cites W1948584817 @default.
• W2892008670 cites W1951289974 @default.
• W2892008670 cites W1994246617 @default.
• W2892008670 cites W1997488488 @default.
• W2892008670 cites W2017387569 @default.
• W2892008670 cites W2020368789 @default.
• W2892008670 cites W2040369854 @default.
• W2892008670 cites W2052003757 @default.
• W2892008670 cites W2053017016 @default.
• W2892008670 cites W2070604790 @default.
• W2892008670 cites W2099471712 @default.
• W2892008670 cites W2129112648 @default.
• W2892008670 cites W2164147879 @default.
• W2892008670 cites W2165633874 @default.
• W2892008670 cites W2296168388 @default.
• W2892008670 cites W2331128040 @default.
• W2892008670 cites W2342877626 @default.
• W2892008670 cites W2523714292 @default.
• W2892008670 cites W2524365899 @default.
• W2892008670 cites W2593414223 @default.
• W2892008670 cites W2593978077 @default.
• W2892008670 cites W2605287558 @default.
• W2892008670 cites W2739748921 @default.
• W2892008670 cites W2740546229 @default.
• W2892008670 cites W2916743882 @default.
• W2892008670 cites W2949290208 @default.
• W2892008670 cites W2950424701 @default.
• W2892008670 cites W2952186111 @default.
• W2892008670 cites W2963073614 @default.
• W2892008670 cites W2963125871 @default.
• W2892008670 cites W2963709863 @default.
• W2892008670 cites W2964121744 @default.
• W2892008670 cites W3102542372 @default.
• W2892008670 cites W603908379 @default.
• W2892008670 doi "https://doi.org/10.48550/arxiv.1809.01372" @default.
• W2892008670 hasPublicationYear "2018" @default.
• W2892008670 type Work @default.
• W2892008670 sameAs 2892008670 @default.
• W2892008670 citedByCount "0" @default.
• W2892008670 crossrefType "posted-content" @default.
• W2892008670 hasAuthorship W2892008670A5027959479 @default.
• W2892008670 hasAuthorship W2892008670A5029588621 @default.
• W2892008670 hasAuthorship W2892008670A5037233582 @default.
• W2892008670 hasAuthorship W2892008670A5052835435 @default.
• W2892008670 hasAuthorship W2892008670A5071037763 @default.
• W2892008670 hasBestOaLocation W28920086701 @default.
• W2892008670 hasConcept C108583219 @default.
• W2892008670 hasConcept C111919701 @default.
• W2892008670 hasConcept C115961682 @default.
• W2892008670 hasConcept C119857082 @default.
• W2892008670 hasConcept C120665830 @default.
• W2892008670 hasConcept C121332964 @default.
• W2892008670 hasConcept C129315195 @default.
• W2892008670 hasConcept C13280743 @default.
• W2892008670 hasConcept C144133560 @default.
• W2892008670 hasConcept C146849305 @default.
• W2892008670 hasConcept C154945302 @default.
• W2892008670 hasConcept C160633673 @default.
• W2892008670 hasConcept C162853370 @default.
• W2892008670 hasConcept C185798385 @default.
• W2892008670 hasConcept C192209626 @default.
• W2892008670 hasConcept C205649164 @default.
• W2892008670 hasConcept C24890656 @default.
• W2892008670 hasConcept C2776436953 @default.
• W2892008670 hasConcept C2779803651 @default.
• W2892008670 hasConcept C2779962950 @default.
• W2892008670 hasConcept C31972630 @default.
• W2892008670 hasConcept C37736160 @default.
• W2892008670 hasConcept C41008148 @default.
• W2892008670 hasConcept C76155785 @default.
• W2892008670 hasConcept C81363708 @default.
• W2892008670 hasConcept C86251818 @default.
• W2892008670 hasConcept C94915269 @default.
• W2892008670 hasConcept C98045186 @default.
• W2892008670 hasConceptScore W2892008670C108583219 @default.
• W2892008670 hasConceptScore W2892008670C111919701 @default.
• W2892008670 hasConceptScore W2892008670C115961682 @default.
• W2892008670 hasConceptScore W2892008670C119857082 @default.
• W2892008670 hasConceptScore W2892008670C120665830 @default.
• W2892008670 hasConceptScore W2892008670C121332964 @default.
• W2892008670 hasConceptScore W2892008670C129315195 @default.
• W2892008670 hasConceptScore W2892008670C13280743 @default.
• W2892008670 hasConceptScore W2892008670C144133560 @default.
• W2892008670 hasConceptScore W2892008670C146849305 @default.
• W2892008670 hasConceptScore W2892008670C154945302 @default.
• W2892008670 hasConceptScore W2892008670C160633673 @default.
• W2892008670 hasConceptScore W2892008670C162853370 @default.

• next