Wrong lesson from GPT paradigm — text is all you need. Other modalities will be incredibly stunning. Can feel it already from DALL-E, which can do something quite useful despite its nascent language ability.
MULTIMODAL AI
-
SD2 Trained on 768 Resolution Still Insufficient
By
–
SD2 is already trained on 768! But still too low res
-

Stable Diffusion Resolution Limitations and Future Improvements
By
–
People don't realize that Stable Diffusion is trained on tiny pics of 512x512px! Which is why you're getting so many artefacts, it's like an AI with really bad vision that needs glasses Reso will exponentially increase: the smallest details can be properly simulated then
-
SD2 768×768 Resolution Still Insufficient for Hand Detail
By
–
Training resolution is 512×512 pics, with SD2 it's 768×768 so that's too small to see hands properly
-
AI-Generated Images: Training on Limited Data Sets
By
–
No she does exist, but it's trained on 20 pics of her and this is AI output
-

Realistic AI-Generated Images That Never Actually Happened
By
–
#36 PROMPTS REALISTAS Vamos con unas cuantas imágenes realistas que nunca fueron tomadas en realidad. Son espectaculares.
-

AI Cinema Generation: Movie Title and Year Prompt Techniques
By
–
#35 CÁMARAS, LUCES, PROMPT Otro truco que suele funcionar bien cuando quieras generar AI CINEMA, es inducir a la IA a una determinada película en base a su título y año. Es decir: "A still from 'Dancing cats' movie (1986)" "A still from 'Rockets to Mars' (2021)"
-

AI as Creative Medium: Photography Without Cameras
By
–
#34 MIRA AL OBJETIVO Y DI "PROMPT" Aquí otra de las fotografías que mi bro sacó y nunca disparó. Tenía la idea en mente de lo que quería, y probando y probando, llegó al resultado. Sí, ya podemos hacer fotografía sin cámaras. La IA es un medio, y la idea plasmada es el fin.
-

Movie-Style Image Generation with AI: Cinema Prompt Techniques
By
–
#32 ¿QUIERES HACER CINE? Para conseguir imágenes espectaculares salidas de una película inexistente, tendrás que: Usar un –ar 3:2 para planos anchos. Incluir en tu prompt descripciones como: "A movie frame of…"
"A movie still from…"
"Cinematographic"
"…"