Extreme Compression of Adaptive Neural Images
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866909846428712960 |
|---|---|
| author | Hoshikawa, Leo Conde, Marcos V. Ohashi, Takeshi Irie, Atsushi |
| author_facet | Hoshikawa, Leo Conde, Marcos V. Ohashi, Takeshi Irie, Atsushi |
| contents | Implicit Neural Representations (INRs) and Neural Fields are a novel paradigm for signal representation, from images and audio to 3D scenes and videos. The fundamental idea is to represent a signal as a continuous and differentiable neural network. This new approach poses new theoretical questions and challenges. Considering a neural image as a 2D image represented as a neural network, we aim to explore novel neural image compression. In this work, we present a novel analysis on compressing neural fields, with focus on images and introduce Adaptive Neural Images (ANI), an efficient neural representation that enables adaptation to different inference or transmission requirements. Our proposed method allows us to reduce the bits-per-pixel (bpp) of the neural image by 8 times, without losing sensitive details or harming fidelity. Our work offers a new framework for developing compressed neural fields. We achieve a new state-of-the-art in terms of PSNR/bpp trade-off thanks to our successful implementation of 4-bit neural representations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_16807 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Extreme Compression of Adaptive Neural Images Hoshikawa, Leo Conde, Marcos V. Ohashi, Takeshi Irie, Atsushi Computer Vision and Pattern Recognition Artificial Intelligence Graphics Multimedia Implicit Neural Representations (INRs) and Neural Fields are a novel paradigm for signal representation, from images and audio to 3D scenes and videos. The fundamental idea is to represent a signal as a continuous and differentiable neural network. This new approach poses new theoretical questions and challenges. Considering a neural image as a 2D image represented as a neural network, we aim to explore novel neural image compression. In this work, we present a novel analysis on compressing neural fields, with focus on images and introduce Adaptive Neural Images (ANI), an efficient neural representation that enables adaptation to different inference or transmission requirements. Our proposed method allows us to reduce the bits-per-pixel (bpp) of the neural image by 8 times, without losing sensitive details or harming fidelity. Our work offers a new framework for developing compressed neural fields. We achieve a new state-of-the-art in terms of PSNR/bpp trade-off thanks to our successful implementation of 4-bit neural representations. |
| title | Extreme Compression of Adaptive Neural Images |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Graphics Multimedia |
| url | https://arxiv.org/abs/2405.16807 |