SoundingActions: Learning How Actions Sound from Narrated Egocentric Videos
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866913305225854976 |
|---|---|
| author | Chen, Changan Ashutosh, Kumar Girdhar, Rohit Harwath, David Grauman, Kristen |
| author_facet | Chen, Changan Ashutosh, Kumar Girdhar, Rohit Harwath, David Grauman, Kristen |
| contents | We propose a novel self-supervised embedding to learn how actions sound from narrated in-the-wild egocentric videos. Whereas existing methods rely on curated data with known audio-visual correspondence, our multimodal contrastive-consensus coding (MC3) embedding reinforces the associations between audio, language, and vision when all modality pairs agree, while diminishing those associations when any one pair does not. We show our approach can successfully discover how the long tail of human actions sound from egocentric video, outperforming an array of recent multimodal embedding techniques on two datasets (Ego4D and EPIC-Sounds) and multiple cross-modal tasks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_05206 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | SoundingActions: Learning How Actions Sound from Narrated Egocentric Videos Chen, Changan Ashutosh, Kumar Girdhar, Rohit Harwath, David Grauman, Kristen Computer Vision and Pattern Recognition Multimedia Sound Audio and Speech Processing We propose a novel self-supervised embedding to learn how actions sound from narrated in-the-wild egocentric videos. Whereas existing methods rely on curated data with known audio-visual correspondence, our multimodal contrastive-consensus coding (MC3) embedding reinforces the associations between audio, language, and vision when all modality pairs agree, while diminishing those associations when any one pair does not. We show our approach can successfully discover how the long tail of human actions sound from egocentric video, outperforming an array of recent multimodal embedding techniques on two datasets (Ego4D and EPIC-Sounds) and multiple cross-modal tasks. |
| title | SoundingActions: Learning How Actions Sound from Narrated Egocentric Videos |
| topic | Computer Vision and Pattern Recognition Multimedia Sound Audio and Speech Processing |
| url | https://arxiv.org/abs/2404.05206 |