Artist-Created Mesh Generation from Raw Observation
This addresses the need for efficient, compatible meshes in commercial graphics pipelines, but it appears incremental as it builds on existing generative models for a specific domain.
The paper tackles the problem of generating artist-style meshes from noisy or incomplete point clouds, such as from LiDAR or RGB-D cameras, by proposing an end-to-end framework that refines inputs and directly produces high-quality meshes, with preliminary results on ShapeNet showing promise.
We present an end-to-end framework for generating artist-style meshes from noisy or incomplete point clouds, such as those captured by real-world sensors like LiDAR or mobile RGB-D cameras. Artist-created meshes are crucial for commercial graphics pipelines due to their compatibility with animation and texturing tools and their efficiency in rendering. However, existing approaches often assume clean, complete inputs or rely on complex multi-stage pipelines, limiting their applicability in real-world scenarios. To address this, we propose an end-to-end method that refines the input point cloud and directly produces high-quality, artist-style meshes. At the core of our approach is a novel reformulation of 3D point cloud refinement as a 2D inpainting task, enabling the use of powerful generative models. Preliminary results on the ShapeNet dataset demonstrate the promise of our framework in producing clean, complete meshes.