Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386071814> ?p ?o ?g. }
Showing items 1 to 70 of
70
with 100 items per page.
- W4386071814 abstract "We present Corgi, a novel method for text-to-image generation. Corgi is based on our proposed shifted diffusion model, which achieves better image embedding generation from input text. Unlike the baseline diffusion model used in DALL-E 2, our method seamlessly encodes prior knowledge of the pre-trained CLIP model in its diffusion process by designing a new initialization distribution and a new transition step of the diffusion. Compared to the strong DALL-E 2 baseline, our method performs better in generating image embedding from the text in terms of both efficiency and effectiveness, resulting in better text-to-image generation. Extensive large-scale experiments are conducted and evaluated in terms of both quantitative measures and human evaluation, indicating a stronger generation ability of our method compared to existing ones. Furthermore, our model enables semi-supervised and language-free training for text-to-image generation, where only part or none of the images in the training dataset have an associated caption. Trained with only 1.7% of the images being captioned, our semi-supervised model obtains FID results comparable to DALL-E 2 on zero-shot text-to-image generation evaluated on MS-COCO. Corgi also achieves new state-of-the-art results across different datasets on downstream language-free text-to-image generation tasks, outperforming the previous method, Lafite, by a large margin." @default.
- W4386071814 created "2023-08-23" @default.
- W4386071814 creator A5006435999 @default.
- W4386071814 creator A5006945148 @default.
- W4386071814 creator A5022713049 @default.
- W4386071814 creator A5028927697 @default.
- W4386071814 creator A5077396200 @default.
- W4386071814 creator A5090601084 @default.
- W4386071814 date "2023-06-01" @default.
- W4386071814 modified "2023-10-16" @default.
- W4386071814 title "Shifted Diffusion for Text-to-image Generation" @default.
- W4386071814 cites W2886641317 @default.
- W4386071814 cites W2962770929 @default.
- W4386071814 cites W2963966654 @default.
- W4386071814 cites W3174194560 @default.
- W4386071814 cites W3176641147 @default.
- W4386071814 doi "https://doi.org/10.1109/cvpr52729.2023.00979" @default.
- W4386071814 hasPublicationYear "2023" @default.
- W4386071814 type Work @default.
- W4386071814 citedByCount "0" @default.
- W4386071814 crossrefType "proceedings-article" @default.
- W4386071814 hasAuthorship W4386071814A5006435999 @default.
- W4386071814 hasAuthorship W4386071814A5006945148 @default.
- W4386071814 hasAuthorship W4386071814A5022713049 @default.
- W4386071814 hasAuthorship W4386071814A5028927697 @default.
- W4386071814 hasAuthorship W4386071814A5077396200 @default.
- W4386071814 hasAuthorship W4386071814A5090601084 @default.
- W4386071814 hasConcept C114466953 @default.
- W4386071814 hasConcept C115961682 @default.
- W4386071814 hasConcept C119857082 @default.
- W4386071814 hasConcept C121332964 @default.
- W4386071814 hasConcept C153180895 @default.
- W4386071814 hasConcept C154945302 @default.
- W4386071814 hasConcept C199360897 @default.
- W4386071814 hasConcept C204321447 @default.
- W4386071814 hasConcept C41008148 @default.
- W4386071814 hasConcept C41608201 @default.
- W4386071814 hasConcept C69357855 @default.
- W4386071814 hasConcept C774472 @default.
- W4386071814 hasConcept C97355855 @default.
- W4386071814 hasConceptScore W4386071814C114466953 @default.
- W4386071814 hasConceptScore W4386071814C115961682 @default.
- W4386071814 hasConceptScore W4386071814C119857082 @default.
- W4386071814 hasConceptScore W4386071814C121332964 @default.
- W4386071814 hasConceptScore W4386071814C153180895 @default.
- W4386071814 hasConceptScore W4386071814C154945302 @default.
- W4386071814 hasConceptScore W4386071814C199360897 @default.
- W4386071814 hasConceptScore W4386071814C204321447 @default.
- W4386071814 hasConceptScore W4386071814C41008148 @default.
- W4386071814 hasConceptScore W4386071814C41608201 @default.
- W4386071814 hasConceptScore W4386071814C69357855 @default.
- W4386071814 hasConceptScore W4386071814C774472 @default.
- W4386071814 hasConceptScore W4386071814C97355855 @default.
- W4386071814 hasFunder F4320306076 @default.
- W4386071814 hasLocation W43860718141 @default.
- W4386071814 hasOpenAccess W4386071814 @default.
- W4386071814 hasPrimaryLocation W43860718141 @default.
- W4386071814 hasRelatedWork W2055709700 @default.
- W4386071814 hasRelatedWork W2072477553 @default.
- W4386071814 hasRelatedWork W2293457016 @default.
- W4386071814 hasRelatedWork W2345283274 @default.
- W4386071814 hasRelatedWork W2368370270 @default.
- W4386071814 hasRelatedWork W2374442885 @default.
- W4386071814 hasRelatedWork W2374512474 @default.
- W4386071814 hasRelatedWork W2789919619 @default.
- W4386071814 hasRelatedWork W3196587153 @default.
- W4386071814 hasRelatedWork W4316511403 @default.
- W4386071814 isParatext "false" @default.
- W4386071814 isRetracted "false" @default.
- W4386071814 workType "article" @default.