Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2403.11556 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917617063690240 |
|---|---|
| author | Zhang, Qianyu Zheng, Bolun Chen, Xinying Chen, Quan Zhu, Zhunjie Wang, Canjin Li, Zongpeng Yan, Chengang |
| author_facet | Zhang, Qianyu Zheng, Bolun Chen, Xinying Chen, Quan Zhu, Zhunjie Wang, Canjin Li, Zongpeng Yan, Chengang |
| contents | Video compression artifacts arise due to the quantization operation in the frequency domain. The goal of video quality enhancement is to reduce compression artifacts and reconstruct a visually-pleasant result. In this work, we propose a hierarchical frequency-based upsampling and refining neural network (HFUR) for compressed video quality enhancement. HFUR consists of two modules: implicit frequency upsampling module (ImpFreqUp) and hierarchical and iterative refinement module (HIR). ImpFreqUp exploits DCT-domain prior derived through implicit DCT transform, and accurately reconstructs the DCT-domain loss via a coarse-to-fine transfer. Consequently, HIR is introduced to facilitate cross-collaboration and information compensation between the scales, thus further refine the feature maps and promote the visual quality of the final output. We demonstrate the effectiveness of the proposed modules via ablation experiments and visualized results. Extensive experiments on public benchmarks show that HFUR achieves state-of-the-art performance for both constant bit rate and constant QP modes. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_11556 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Hierarchical Frequency-based Upsampling and Refining for Compressed Video Quality Enhancement Zhang, Qianyu Zheng, Bolun Chen, Xinying Chen, Quan Zhu, Zhunjie Wang, Canjin Li, Zongpeng Yan, Chengang Image and Video Processing Computer Vision and Pattern Recognition Video compression artifacts arise due to the quantization operation in the frequency domain. The goal of video quality enhancement is to reduce compression artifacts and reconstruct a visually-pleasant result. In this work, we propose a hierarchical frequency-based upsampling and refining neural network (HFUR) for compressed video quality enhancement. HFUR consists of two modules: implicit frequency upsampling module (ImpFreqUp) and hierarchical and iterative refinement module (HIR). ImpFreqUp exploits DCT-domain prior derived through implicit DCT transform, and accurately reconstructs the DCT-domain loss via a coarse-to-fine transfer. Consequently, HIR is introduced to facilitate cross-collaboration and information compensation between the scales, thus further refine the feature maps and promote the visual quality of the final output. We demonstrate the effectiveness of the proposed modules via ablation experiments and visualized results. Extensive experiments on public benchmarks show that HFUR achieves state-of-the-art performance for both constant bit rate and constant QP modes. |
| title | Hierarchical Frequency-based Upsampling and Refining for Compressed Video Quality Enhancement |
| topic | Image and Video Processing Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2403.11556 |