High-Frequency Enhanced Hybrid Neural Representation for Video Compression
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866918004937195520 |
|---|---|
| author | Yu, Li Li, Zhihui Xiao, Jimin Gabbouj, Moncef |
| author_facet | Yu, Li Li, Zhihui Xiao, Jimin Gabbouj, Moncef |
| contents | Neural Representations for Videos (NeRV) have simplified the video codec process and achieved swift decoding speeds by encoding video content into a neural network, presenting a promising solution for video compression. However, existing work overlooks the crucial issue that videos reconstructed by these methods lack high-frequency details. To address this problem, this paper introduces a High-Frequency Enhanced Hybrid Neural Representation Network. Our method focuses on leveraging high-frequency information to improve the synthesis of fine details by the network. Specifically, we design a wavelet high-frequency encoder that incorporates Wavelet Frequency Decomposer (WFD) blocks to generate high-frequency feature embeddings. Next, we design the High-Frequency Feature Modulation (HFM) block, which leverages the extracted high-frequency embeddings to enhance the fitting process of the decoder. Finally, with the refined Harmonic decoder block and a Dynamic Weighted Frequency Loss, we further reduce the potential loss of high-frequency information. Experiments on the Bunny and UVG datasets demonstrate that our method outperforms other methods, showing notable improvements in detail preservation and compression performance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2411_06685 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | High-Frequency Enhanced Hybrid Neural Representation for Video Compression Yu, Li Li, Zhihui Xiao, Jimin Gabbouj, Moncef Computer Vision and Pattern Recognition Artificial Intelligence Image and Video Processing Neural Representations for Videos (NeRV) have simplified the video codec process and achieved swift decoding speeds by encoding video content into a neural network, presenting a promising solution for video compression. However, existing work overlooks the crucial issue that videos reconstructed by these methods lack high-frequency details. To address this problem, this paper introduces a High-Frequency Enhanced Hybrid Neural Representation Network. Our method focuses on leveraging high-frequency information to improve the synthesis of fine details by the network. Specifically, we design a wavelet high-frequency encoder that incorporates Wavelet Frequency Decomposer (WFD) blocks to generate high-frequency feature embeddings. Next, we design the High-Frequency Feature Modulation (HFM) block, which leverages the extracted high-frequency embeddings to enhance the fitting process of the decoder. Finally, with the refined Harmonic decoder block and a Dynamic Weighted Frequency Loss, we further reduce the potential loss of high-frequency information. Experiments on the Bunny and UVG datasets demonstrate that our method outperforms other methods, showing notable improvements in detail preservation and compression performance. |
| title | High-Frequency Enhanced Hybrid Neural Representation for Video Compression |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Image and Video Processing |
| url | https://arxiv.org/abs/2411.06685 |