Universal End-to-End Neural Network for Lossy Image Compression
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866909310811897856 |
|---|---|
| author | Arezki, Bouzid Feng, Fangchen Mokraoui, Anissa |
| author_facet | Arezki, Bouzid Feng, Fangchen Mokraoui, Anissa |
| contents | This paper presents variable bitrate lossy image compression using a VAE-based neural network. An adaptable image quality adjustment strategy is proposed. The key innovation involves adeptly adjusting the input scale exclusively during the inference process, resulting in an exceptionally efficient rate-distortion mechanism. Through extensive experimentation, across diverse VAE-based compression architectures (CNN, ViT) and training methodologies (MSE, SSIM), our approach exhibits remarkable universality. This success is attributed to the inherent generalization capacity of neural networks. Unlike methods that adjust model architecture or loss functions, our approach emphasizes simplicity, reducing computational complexity and memory requirements. The experiments not only highlight the effectiveness of our approach but also indicate its potential to drive advancements in variable-rate neural network lossy image compression methodologies. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_06586 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Universal End-to-End Neural Network for Lossy Image Compression Arezki, Bouzid Feng, Fangchen Mokraoui, Anissa Image and Video Processing This paper presents variable bitrate lossy image compression using a VAE-based neural network. An adaptable image quality adjustment strategy is proposed. The key innovation involves adeptly adjusting the input scale exclusively during the inference process, resulting in an exceptionally efficient rate-distortion mechanism. Through extensive experimentation, across diverse VAE-based compression architectures (CNN, ViT) and training methodologies (MSE, SSIM), our approach exhibits remarkable universality. This success is attributed to the inherent generalization capacity of neural networks. Unlike methods that adjust model architecture or loss functions, our approach emphasizes simplicity, reducing computational complexity and memory requirements. The experiments not only highlight the effectiveness of our approach but also indicate its potential to drive advancements in variable-rate neural network lossy image compression methodologies. |
| title | Universal End-to-End Neural Network for Lossy Image Compression |
| topic | Image and Video Processing |
| url | https://arxiv.org/abs/2409.06586 |