DORSal: Diffusion for Object-centric Representations of Scenes et al
This addresses the trade-off between rendering quality and scalability in 3D scene representation learning, enabling better generalization and editing for applications like virtual reality or autonomous driving.
The paper tackles the problem of rendering high-fidelity novel views in 3D scene understanding while retaining object-level editing capabilities, and shows that DORSal improves upon existing approaches on synthetic and real-world datasets.
Recent progress in 3D scene understanding enables scalable learning of representations across large datasets of diverse scenes. As a consequence, generalization to unseen scenes and objects, rendering novel views from just a single or a handful of input images, and controllable scene generation that supports editing, is now possible. However, training jointly on a large number of scenes typically compromises rendering quality when compared to single-scene optimized models such as NeRFs. In this paper, we leverage recent progress in diffusion models to equip 3D scene representation learning models with the ability to render high-fidelity novel views, while retaining benefits such as object-level scene editing to a large degree. In particular, we propose DORSal, which adapts a video diffusion architecture for 3D scene generation conditioned on frozen object-centric slot-based representations of scenes. On both complex synthetic multi-object scenes and on the real-world large-scale Street View dataset, we show that DORSal enables scalable neural rendering of 3D scenes with object-level editing and improves upon existing approaches.