From Noise to Latent: Generating Gaussian Latents for INR-Based Image Compression
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866914150595166208 |
|---|---|
| author | Lin, Chaoyi Wu, Yaojun Li, Yue Li, Junru Zhang, Kai Zhang, Li |
| author_facet | Lin, Chaoyi Wu, Yaojun Li, Yue Li, Junru Zhang, Kai Zhang, Li |
| contents | Recent implicit neural representation (INR)-based image compression methods have shown competitive performance by overfitting image-specific latent codes. However, they remain inferior to end-to-end (E2E) compression approaches due to the absence of expressive latent representations. On the other hand, E2E methods rely on transmitting latent codes and requiring complex entropy models, leading to increased decoding complexity. Inspired by the normalization strategy in E2E codecs where latents are transformed into Gaussian noise to demonstrate the removal of spatial redundancy, we explore the inverse direction: generating latents directly from Gaussian noise. In this paper, we propose a novel image compression paradigm that reconstructs image-specific latents from a multi-scale Gaussian noise tensor, deterministically generated using a shared random seed. A Gaussian Parameter Prediction (GPP) module estimates the distribution parameters, enabling one-shot latent generation via reparameterization trick. The predicted latent is then passed through a synthesis network to reconstruct the image. Our method eliminates the need to transmit latent codes while preserving latent-based benefits, achieving competitive rate-distortion performance on Kodak and CLIC dataset. To the best of our knowledge, this is the first work to explore Gaussian latent generation for learned image compression. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_08009 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | From Noise to Latent: Generating Gaussian Latents for INR-Based Image Compression Lin, Chaoyi Wu, Yaojun Li, Yue Li, Junru Zhang, Kai Zhang, Li Image and Video Processing Computer Vision and Pattern Recognition Recent implicit neural representation (INR)-based image compression methods have shown competitive performance by overfitting image-specific latent codes. However, they remain inferior to end-to-end (E2E) compression approaches due to the absence of expressive latent representations. On the other hand, E2E methods rely on transmitting latent codes and requiring complex entropy models, leading to increased decoding complexity. Inspired by the normalization strategy in E2E codecs where latents are transformed into Gaussian noise to demonstrate the removal of spatial redundancy, we explore the inverse direction: generating latents directly from Gaussian noise. In this paper, we propose a novel image compression paradigm that reconstructs image-specific latents from a multi-scale Gaussian noise tensor, deterministically generated using a shared random seed. A Gaussian Parameter Prediction (GPP) module estimates the distribution parameters, enabling one-shot latent generation via reparameterization trick. The predicted latent is then passed through a synthesis network to reconstruct the image. Our method eliminates the need to transmit latent codes while preserving latent-based benefits, achieving competitive rate-distortion performance on Kodak and CLIC dataset. To the best of our knowledge, this is the first work to explore Gaussian latent generation for learned image compression. |
| title | From Noise to Latent: Generating Gaussian Latents for INR-Based Image Compression |
| topic | Image and Video Processing Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2511.08009 |