Asking a single model to simultaneously handle prompt understanding, scene composition, interaction reasoning, identity preservation, and rendering for multi-person scenes is too much; separating planning from rendering improves results.
The R2-Canvas paper separates image generation into an Architect (planning composition) and Artist (rendering pixels), mimicking how human artists approach scene creation rather than doing everything in one pass. ✦ AI generated
Fatih Porikli · The TWIML AI Podcast · 2026-08-12 · original ↗
starts at this moment · 13:39
“are we in doing so would we be asking the models to do too many things at once?”
maybe we are also asking a single model to solve too many difficult problems at once. Think about what happens when you generate a scene with several people. The model has to understand the prompt, decide how many people should appear, determine where they should be placed like the composition, reason about their interactions, preserve the identity of the person... and finally render everything together in a single process. Maybe this is too much. Instead of asking one model to do everything what if we separated planning from rendering similar to what human artists might do.
verbatim transcript · starts at 13:39
13:19objective. Optimization at the right data is definitely very critical. That is what this paper is talking about. But we may also kind of think that maybe I mean these are models genetic models. Maybe we are also asking a single model to solve too many difficult problems at once. >> That's kind of asking the question could we possibly optimize for all the attributes that we care about? Are we a
13:46in doing so would we be asking the models to do too many things at once? >> That's a good point. Maybe we shouldn't kind of um and also I can give you an example. Think about what happens when you generate a scene with several people. Going back to that core example now the model has to understand the prompt but what I'm asking to model and then decide how many people should
14:09appear uh determine where they should be placed like the composition of the scene reason about their interactions because if there's a person if there's another person most likely there there is some connection preserve and you also I didn't mention about this thing identity of the person we can give okay this is my daughter this is my son and I want them to be in the picture not like any
14:31random person and finally render everything together in all a single process. uh maybe this is too much and that's what we explored in the other paper R2 CAM paper instead of like asking one model to do everything what if we separated for instance planning from rendering similar to what human artists might do uh and you know that kind of that kind of led us into this uh
15:01paper >> before we dive into Artican I I thinking about that comment applied to disco and again this idea about different attributes possibly a thing to keep in mind is how we've matured the way we think about like using model routers. Uh so um you know maybe you have you know a suite of models for the different attributes that you care about and then when your prompt comes in you've got
- ·Single model must understand prompts, compose scenes, reason interactions, and preserve identities all at once.
- ·Multi-person scenes add extreme complexity — deciding who appears, where they go, and how they interact.
- ·Human artists separate planning from rendering; current models try to do everything in one pass.
- ·R2-Canvas introduces Architect (planning composition) and Artist (rendering pixels).
- ·Mimics human creative process — think first, then draw.
- ·Separated approach improves results for complex multi-person scenes.