Matches in SemOpenAlex for { <https://semopenalex.org/work/W2988200020> ?p ?o ?g. }
Showing items 1 to 96 of
96
with 100 items per page.
- W2988200020 abstract "Learning how objects sound from video is challenging, since they often heavily overlap in a single audio channel. Current methods for visually-guided audio source separation sidestep the issue by training with artificially mixed video clips, but this puts unwieldy restrictions on training data collection and may even prevent learning the properties of true mixed sounds. We introduce a co-separation training paradigm that permits learning object-level sounds from unlabeled multi-source videos. Our novel training objective requires that the deep neural network's separated audio for similar-looking objects be consistently identifiable, while simultaneously reproducing accurate video-level audio tracks for each source training pair. Our approach disentangles sounds in realistic test videos, even in cases where an object was not observed individually during training. We obtain state-of-the-art results on visually-guided audio source separation and audio denoising for the MUSIC, AudioSet, and AV-Bench datasets." @default.
- W2988200020 created "2019-11-22" @default.
- W2988200020 creator A5012765543 @default.
- W2988200020 creator A5032081267 @default.
- W2988200020 date "2019-10-01" @default.
- W2988200020 modified "2023-10-14" @default.
- W2988200020 title "Co-Separating Sounds of Visual Objects" @default.
- W2988200020 cites W1931249336 @default.
- W2988200020 cites W2031647436 @default.
- W2988200020 cites W2034424181 @default.
- W2988200020 cites W2039844283 @default.
- W2988200020 cites W2065760681 @default.
- W2988200020 cites W2088049833 @default.
- W2988200020 cites W2098950531 @default.
- W2988200020 cites W2105582566 @default.
- W2988200020 cites W2120847449 @default.
- W2988200020 cites W2123649031 @default.
- W2988200020 cites W2150415460 @default.
- W2988200020 cites W2157244733 @default.
- W2988200020 cites W2167353662 @default.
- W2988200020 cites W2171819471 @default.
- W2988200020 cites W2194775991 @default.
- W2988200020 cites W2221409856 @default.
- W2988200020 cites W2460742184 @default.
- W2988200020 cites W2562551209 @default.
- W2988200020 cites W2593116425 @default.
- W2988200020 cites W2644497536 @default.
- W2988200020 cites W2651884604 @default.
- W2988200020 cites W2660943524 @default.
- W2988200020 cites W2737258237 @default.
- W2988200020 cites W2962866211 @default.
- W2988200020 cites W2963066677 @default.
- W2988200020 cites W2963082324 @default.
- W2988200020 cites W2963680395 @default.
- W2988200020 cites W2963807156 @default.
- W2988200020 cites W2964171275 @default.
- W2988200020 cites W2964229681 @default.
- W2988200020 cites W2964345931 @default.
- W2988200020 cites W2981851635 @default.
- W2988200020 cites W4289665794 @default.
- W2988200020 doi "https://doi.org/10.1109/iccv.2019.00398" @default.
- W2988200020 hasPublicationYear "2019" @default.
- W2988200020 type Work @default.
- W2988200020 sameAs 2988200020 @default.
- W2988200020 citedByCount "156" @default.
- W2988200020 countsByYear W29882000202019 @default.
- W2988200020 countsByYear W29882000202020 @default.
- W2988200020 countsByYear W29882000202021 @default.
- W2988200020 countsByYear W29882000202022 @default.
- W2988200020 countsByYear W29882000202023 @default.
- W2988200020 crossrefType "proceedings-article" @default.
- W2988200020 hasAuthorship W2988200020A5012765543 @default.
- W2988200020 hasAuthorship W2988200020A5032081267 @default.
- W2988200020 hasBestOaLocation W29882000202 @default.
- W2988200020 hasConcept C121332964 @default.
- W2988200020 hasConcept C153294291 @default.
- W2988200020 hasConcept C154945302 @default.
- W2988200020 hasConcept C2776864781 @default.
- W2988200020 hasConcept C2777211547 @default.
- W2988200020 hasConcept C2781238097 @default.
- W2988200020 hasConcept C28490314 @default.
- W2988200020 hasConcept C3017588708 @default.
- W2988200020 hasConcept C31972630 @default.
- W2988200020 hasConcept C41008148 @default.
- W2988200020 hasConcept C49774154 @default.
- W2988200020 hasConcept C50644808 @default.
- W2988200020 hasConceptScore W2988200020C121332964 @default.
- W2988200020 hasConceptScore W2988200020C153294291 @default.
- W2988200020 hasConceptScore W2988200020C154945302 @default.
- W2988200020 hasConceptScore W2988200020C2776864781 @default.
- W2988200020 hasConceptScore W2988200020C2777211547 @default.
- W2988200020 hasConceptScore W2988200020C2781238097 @default.
- W2988200020 hasConceptScore W2988200020C28490314 @default.
- W2988200020 hasConceptScore W2988200020C3017588708 @default.
- W2988200020 hasConceptScore W2988200020C31972630 @default.
- W2988200020 hasConceptScore W2988200020C41008148 @default.
- W2988200020 hasConceptScore W2988200020C49774154 @default.
- W2988200020 hasConceptScore W2988200020C50644808 @default.
- W2988200020 hasLocation W29882000201 @default.
- W2988200020 hasLocation W29882000202 @default.
- W2988200020 hasOpenAccess W2988200020 @default.
- W2988200020 hasPrimaryLocation W29882000201 @default.
- W2988200020 hasRelatedWork W2074916782 @default.
- W2988200020 hasRelatedWork W2271369634 @default.
- W2988200020 hasRelatedWork W230091440 @default.
- W2988200020 hasRelatedWork W2330333072 @default.
- W2988200020 hasRelatedWork W2350550760 @default.
- W2988200020 hasRelatedWork W2393726419 @default.
- W2988200020 hasRelatedWork W3137890128 @default.
- W2988200020 hasRelatedWork W4245955731 @default.
- W2988200020 hasRelatedWork W578794879 @default.
- W2988200020 hasRelatedWork W2625296515 @default.
- W2988200020 isParatext "false" @default.
- W2988200020 isRetracted "false" @default.
- W2988200020 magId "2988200020" @default.
- W2988200020 workType "article" @default.