This repository provides a Gradio-based demo for IDT (Intrinsic Diffusion Transformer), a model designed for intrinsic image decomposition with interactive relighting-style editing.

Given a video or a set of images, the demo predicts per-frame intrinsic components and allows users to interactively modify shading color and intensity, enabling relighting-like effects without re-running inference.

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support