SealD-NeRF: Interactive Pixel-Level Editing for Dynamic Scenes by Neural Radiance Fields

arXiv:2402.13510v1 Announce Type: new
Abstract: The widespread adoption of implicit neural representations, especially Neural Radiance Fields (NeRF), highlights a growing need for editing capabilities in implicit 3D models, essential for tasks like scene post-processing and 3D content creation. Despite previous efforts in NeRF editing, challenges remain due to limitations in editing flexibility and quality. The key issue is developing a neural representation that supports local edits for real-time updates. Current NeRF editing methods, offering pixel-level adjustments or detailed geometry and …

3d models abstract adoption arxiv capabilities challenges cs.cv dynamic editing fields highlights implicit neural representations interactive limitations nerf neural radiance fields pixel post-processing processing tasks type