Welcome to the ARTSMARTAI roadmap!

صوره للإحتفاء بيوم المعلم العماني

kartenlegen

Präsident obama mit Merkel in der U-bahn

A sloth hanging from a tree with a rifle hanging from its shoulders with a fool moon in the background.

realiser une image de tempete avec pompiers du GSCF dos du pompier inscription GSCF

Wireframe site rencontre de chaussettes 

Black man eating ice cream

air force 1  low billie eilish personnalisé

codeformer or similler is not worthy but adetailer or something similler can be add for fixing face, hands and feet. Also, a refiner is must required for final output. 

Able to Train an AI for a character (to be saved and use for future prompts)
Able to Train the AI by uploading multiple similar images - basically user generated "Artistic Style"

The above can add up to the features where artist and prompt engineer can sell their character and Artistic Style also

ART SCULPTUR PONTIGNY ABBAYE

Hey All.

it would be great, if I could place in the image-to-image edit tool two different masks and can adjust the strength of the AI for both separatelly. For example for the face - one value, and for the body - another value. So I would be more flexible to create AI-content.

Thanks

man with demantia on a trainstation wearing his pyjama waiting for the tram in zurich (switzerland)

Generate a stage for a modern version of Joan of Arc

Allow users to edit image outputs without having to start from scratch.

Two current ideas.
Allow the user to train the AI by rating their satisfaction with the image output from 0 to 10. The system could create tags (variable identifiers) for each adjustment within that session. Once a style or character is like the system can save and name/identify the variables. These tags could be added to the prompt and saved for later use. Tags can be created for the image style, character, clothing, or objects within the image.

When users select ArtSmart generated images from "my gallery" as a reference image an additional box will appear offering the option to keep or change. The user could then provide text prompts for the specific variables they would like to adjust within that image.
example keep: background, eye colors, clothing, and hairstyle
example change: skin tone to F6E311, adjust pose to standing, character is smiling

Have a gridded row column system to allow select detailed edits to images.

This will allow more detailed configuration.

Process:
A user would select a finished image. The system would then align the image to the grid. The user could then select cells on the grid and provide prompts to that portion of the image only.

ie user would state "change eye color in cell 5 row 7."

This would allow changes without affecting the rest of the image. It would also reduce computing power as a complete render is not required for changes as well as requesting multiple outputs for one image.

Faire une affiche graphique pour une opération de don de poules aux habitants pour réduire les déchets

lorraine selmer cooking couscous

Person sitting in german office writing letters with chat gbt

Bild von Christian Knecht und einer KI

personnel de développement

Please add more models and modes that enable to create illustrations (like coloring pages, cartoon pages) more easily.

When using NUM Outputs > 1 it would be great if you could enter a stepping for variable parameters, like e.g. age of a person: "Age n +20" OR speed of Wind n=25 (light breeze) +25 (n=100 Stormwind)

Plane Modellbahnanlage mit dem Gleis h0 PIKO eine Modellbahn mit spitzkehre, 2 Gleisen, schattenbahnhof, tunnel mit steigung 2,5%, kleinster Radius im Tunnel 550mm, kleinster sichtbarer Radius 750 mm, Länge 10m, breite 2,3 m

Male ein Logo mit einer Ente