ReVersion: Diffusion-Based Relation Inversion from Images
ReVersion "Michael Jackson <R> wall" "cat <R> canvas" "Spiderman <R> wall" "hamburger <R> vase" "cat <R> building" exemplar images <R> ReVersion "cat <R> cat" "panda <R> panda" "otter <R> otter" "rabbit <R> child" "Batman <R> Batman" exemplar images <R> ReVersion "Spiderman <R> basket" "hamster <R> paper bag" "sea <R> cup" "panda <R> pot" "rabbit <R> cup"Fig. 1.We propose a new task, Relation Inversion: Given a few exemplar images, where a relation co-exists in every image, we aim to find a relation prompt R to capture this interaction, and apply the relation to new entities to synthesize new scenes.The above images are generated by our ReVersion Framework.
Authors
- Kelvin C. K. Chan (ORCID: https://orcid.org/0000-0002-5456-8991)
- Yuming Jiang (ORCID: https://orcid.org/0000-0001-7653-4015)
- Ziwei Liu (ORCID: https://orcid.org/0000-0002-4220-5958)
- Ziqi Huang (ORCID: https://orcid.org/0000-0001-8008-5873)
- Tianxing Wu (ORCID: https://orcid.org/0000-0001-7345-0254)
Institutions
- Nanyang Technological University (SG)
Publication Details
- Published
- 2024-12-03
- DOI
- https://doi.org/10.1145/3680528.3687658
- Citations
- 22
- Primary Topic
- Multimodal Machine Learning Applications
- Type
- article
- Field-Weighted Citation Impact
- 5.48
Funders
- Ministry of Education - Singapore
- Ministry of Education, India
- Bộ Giáo dục và Ðào tạo