NeRFCodec: Neural Feature Compression Meets Neural Radiance Fields for Memory-Efficient Scene Representation

arXiv:2404.02185v1 Announce Type: new
Abstract: The emergence of Neural Radiance Fields (NeRF) has greatly impacted 3D scene modeling and novel-view synthesis. As a kind of visual media for 3D scene representation, compression with high rate-distortion performance is an eternal target. Motivated by advances in neural compression and neural field representation, we propose NeRFCodec, an end-to-end NeRF compression framework that integrates non-linear transform, quantization, and entropy coding for memory-efficient scene representation. Since training a non-linear transform directly on a large scale …

abstract advances arxiv compression cs.cv cs.gr eess.iv emergence feature fields kind media memory modeling nerf neural compression neural radiance fields novel performance rate representation synthesis type view visual