From Sketch to Photorealistic Facades—No As-Built Model Needed

Kari Jaaskelainen

14 Jan 2026 — 1 min read

Renovate, don’t demolish

Producing facade proposals that respect existing structures usually requires painstaking as-built modeling. This research shows a faster path: turn a rough sketch and a short text prompt into a photorealistic renovation concept.

Spot changes: a vision–language model reads your sketch and words to mark where to add or modify elements.
Draft details: a diffusion model generates detailed new components and merges them with the original outline via inpainting.
Make it real: ControlNet refines everything into a consistent, photorealistic image.

Across datasets and real industrial buildings, the framework preserved structure while improving facade quality, enabling rapid iteration and clearer client communication—without building a full as-built model first.

Read more: https://arxiv.org/abs/2601.08531v1

Paper: https://arxiv.org/abs/2601.08531v1

Register: https://www.AiFeta.com

AI architecture adaptive-reuse generative-design computer-vision diffusion-models ControlNet VLM sustainability

Meet GRACE: a moral governor for safer, more transparent AI

AI agents are getting powerful—so how do we make sure they do the right thing, not just the effective thing? Meet GRACE, a reason-based moral governor that keeps AI behavior aligned with human norms by separating moral reasoning from goal-driven decision-making. * Moral Module: uses deontic logic and explicit reasons

Why some fine-tuned LLMs miss phishing—and how to fix it

Not all fine-tuned LLMs spot phishing equally. A new study tests Llama 3.1 8B, Gemma 2 9B, and Mistral on high-stakes phishing detection—and uses SHAP and mechanistic interpretability to reveal why models do (or don’t) generalize. * Architecture × data diversity matters: Gemma 2 9B hits state-of-the-art performance (F1

AI that explains itself—by following the science

AI that explains itself—by following the science Black-box AI can be powerful, but it often can’t tell us why it made a decision. Concept Bottleneck Models (CBMs) try to fix that by predicting human-understandable concepts first, then the final answer. The catch: standard CBMs ignore domain-specific cause-and-effect and

Meet GenomAgent: A Team of AI Specialists for Smarter Genomics Q&A

TL;DR: Finding reliable answers in genomics is hard. GenomAgent turns one big AI into a coordinating team of specialists—and beats the current leader by 12% on a key benchmark. Genomic facts live across many databases. Standard chatbots struggle because they can’t flexibly query those sources. GeneGPT added