Marigold V2: one diffusion step for depth, normals, albedo—sees through glass
AntonObukhov1 · x · 2026-09-15
Anton Obukhov's team released Marigold V2, which extracts depth, see-through depth (finally handling glass), surface normals, and albedo from a single photo—each with just one diffusion step. Live on Hugging Face Spaces.
More from Multimodal
- One image turned into a seamless 4-part carousel with Midjourney and ChatGPT — umesh_ai · 2026-09-15
- Flora ships a video editor MCP, letting the Astra agent drive video editing — round · 2026-09-15
- Open-source skill turns one sentence into a single-file web 3D documentary via Claude Code — xiaohu · 2026-09-15
- Hyper3D launches MCP service, letting Codex auto-build 3D models into explainer sites — xiaohu · 2026-09-15
- Kling AI takes its video models to TIFF 2026, pitching AI video for professional film production — aziz4ai · 2026-09-15
- Bringing a Rubens painting to life with ComfyUI — workflow details shared — VictorVisuals · 2026-09-15