Best Prompts for Text-to-Image Models and How to Find Them
Contenu
- Titre
- Best Prompts for Text-to-Image Models and How to Find Them
- Date de soumission
- 20 février 2023, 12:04:40 +00:00
- Est référencé par
- YY7585F3
- Résumé
- Recent progress in generative models, especially in text-guided diffusion models, has enabled the production of aesthetically-pleasing imagery resembling the works of professional human artists. However, one has to carefully compose the textual description, called the prompt, and augment it with a set of clarifying keywords. Since aesthetics are challenging to evaluate computationally, human feedback is needed to determine the optimal prompt formulation and keyword combination. In this paper, we present a human-in-the-loop approach to learning the most useful combination of prompt keywords using a genetic algorithm. We also show how such an approach can improve the aesthetic appeal of images depicting the same descriptions.
- Date
- 2022-12-02
- Source
- arXiv.org
- is compiled by
- Lucky Semiosis
- is in semantic relation with
- computer science
- Human-Computer Interaction
- H.5.2
- computer science
- Computation and Language
- computer science
- Computer Vision and Pattern Recognition
- H.3.3
- Complexité
-
575
- Date de modification
- 8 septembre 2023, 06:54:15 +00:00
- Détails de la complexité
- Physique,1,,,,,53,53
- Physique,2,,,,,43,86
- Actant,2,,,,,4,8
- Concept,1,,,,,52,52
- Concept,2,,,,,47,94
- Rapport,1,1,Physique,Concept,properties,52,52
- Rapport,1,1,Physique,Physique,values,52,52
- Rapport,1,1,Physique,Actant,dcterms:creator,3,3
- Rapport,2,2,Actant,Concept,properties,19,38
- Rapport,2,2,Actant,Physique,values,19,38
- Rapport,1,1,Physique,Actant,cito:isCompiledBy,1,1
- Rapport,1,1,Physique,Concept,skos:semanticRelation,6,6
- Rapport,2,2,Concept,Concept,properties,18,36
- Rapport,2,2,Concept,Physique,values,18,36
- Rapport,1,1,Physique,Physique,media,2,2
- Rapport,2,2,Physique,Concept,properties,4,8
- Rapport,2,2,Physique,Physique,values,4,8
- Rapport,1,1,Physique,Physique,uri,2,2
- Totaux de la complexité
- Physique,2,1,2,96,139
- Actant,1,2,2,4,8
- Concept,2,1,2,99,146
- Rapport,13,1,2,200,282
- Existence,18,1,2,399,575
- Collections
- Zotero
Ressources liées
| Titre | Classe |
|---|---|
| Note : TN75R2RF | Note |
Annotations
| Value | Purpose |
|---|---|
|
Zotero automatique tagger a taggé le document Best Prompts for Text-to-Image Models and How to Find Them avec le tag computer science
|
classifying |
| Target selector | Selector type |
|---|---|
| o:Item |
| Value | Purpose |
|---|---|
|
Zotero automatique tagger a taggé le document Best Prompts for Text-to-Image Models and How to Find Them avec le tag Human-Computer Interaction
|
classifying |
| Target selector | Selector type |
|---|---|
| o:Item |
| Value | Purpose |
|---|---|
|
Zotero automatique tagger a taggé le document Best Prompts for Text-to-Image Models and How to Find Them avec le tag H.5.2
|
classifying |
| Target selector | Selector type |
|---|---|
| o:Item |
| Value | Purpose |
|---|---|
|
Zotero automatique tagger a taggé le document Best Prompts for Text-to-Image Models and How to Find Them avec le tag Computation and Language
|
classifying |
| Target selector | Selector type |
|---|---|
| o:Item |
| Value | Purpose |
|---|---|
|
Zotero automatique tagger a taggé le document Best Prompts for Text-to-Image Models and How to Find Them avec le tag Computer Vision and Pattern Recognition
|
classifying |
| Target selector | Selector type |
|---|---|
| o:Item |
| Value | Purpose |
|---|---|
|
Zotero automatique tagger a taggé le document Best Prompts for Text-to-Image Models and How to Find Them avec le tag H.3.3
|
classifying |
| Target selector | Selector type |
|---|---|
| o:Item |