DreamPolisher: Towards High-Quality Text-to-3D Generation via Geometric Diffusion
This addresses the challenge of ensuring cross-view consistency and textural richness in text-to-3D generation, which is important for applications in 3D content creation, but it appears incremental as it builds on existing Gaussian Splatting and ControlNet techniques.
The paper tackles the problem of generating view-consistent and detailed 3D objects from text descriptions by proposing DreamPolisher, a two-stage Gaussian Splatting method with geometric guidance, which achieves realistic 3D generation that aligns closely with textual semantics across diverse object categories.
We present DreamPolisher, a novel Gaussian Splatting based method with geometric guidance, tailored to learn cross-view consistency and intricate detail from textual descriptions. While recent progress on text-to-3D generation methods have been promising, prevailing methods often fail to ensure view-consistency and textural richness. This problem becomes particularly noticeable for methods that work with text input alone. To address this, we propose a two-stage Gaussian Splatting based approach that enforces geometric consistency among views. Initially, a coarse 3D generation undergoes refinement via geometric optimization. Subsequently, we use a ControlNet driven refiner coupled with the geometric consistency term to improve both texture fidelity and overall consistency of the generated 3D asset. Empirical evaluations across diverse textual prompts spanning various object categories demonstrate the efficacy of DreamPolisher in generating consistent and realistic 3D objects, aligning closely with the semantics of the textual instructions.