Research map / Video / Video segmentation and tracking
Unidirectional Video Denoising by Mimicking Backward Recurrent Modules with Look-Ahead Forward Ones
Junyi Li, Xiaohe Wu, Zhenxing Niu, Wangmeng Zuo
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add look-ahead future frames (FloRNN vs ForwardRNN) | FloRNN · Set8 (Gaussian, avg over sigma 10-50) | PSNR: 33.12 -> 33.55 | Improved |
| use bidirectional recurrence (BiRNN) instead of look-ahead FloRNN | FloRNN · Set8 (Gaussian, avg over sigma 10-50) | PSNR: 33.55 -> 33.74 | Improved |
| use forward warping instead of backward warping for look-ahead alignment (ablation) | FloRNN · Set8 (Gaussian, avg over sigma 10-50) | PSNR: 33.44 -> 33.45 | Improved |
| add border enlargement to forward warping (ablation) | FloRNN · Set8 (Gaussian, avg over sigma 10-50) | PSNR: 33.45 -> 33.55 | Improved |
| increase near-future frames k from 0 to 3 (ablation) | FloRNN · Set8 (Gaussian, avg over sigma 10-50) | PSNR: 33.12 -> 33.55 | Improved |
| increase near-future frames k beyond 3 to 5 (ablation) | FloRNN · Set8 (Gaussian, avg over sigma 10-50) | PSNR: 33.55 -> 33.51 | No clear effect |
| add knowledge distillation from pretrained BiRNN (ablation) | FloRNN · Set8 (Gaussian, avg over sigma 10-50) | PSNR: 33.55 -> 33.53 | No clear effect |
| add flow-guided deformable alignment and second-order propagation (FloRNN++) | FloRNN · Set8 (Gaussian, avg over sigma 10-50) | PSNR: 33.55 -> 33.72 | Improved |
| add downsample/upsample layers to BasicVSR (ablation) | BasicVSR · Set8 (Gaussian, avg over sigma 10-50) | PSNR: 33.81 -> 31.66 | Hurt |
| add downsample/upsample layers to BasicVSR++ (ablation) | BasicVSR++ · Set8 (Gaussian, avg over sigma 10-50) | PSNR: 34.11 -> 32.81 | Hurt |
| FloRNN vs PaCNet on clipped Gaussian DAVIS | FloRNN · DAVIS (clipped Gaussian) | PSNR: 35.73 -> 36.49 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- FastDVDNet — FastDVDnet: Towards Real-Time Deep Video Denoising Without Flow Estimation (CVPR 2020) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- ReMoNet: Recurrent Multi-Output Network for Efficient Video Denoising (AAAI 2022 · similarity 0.88)
- Deep Recurrent Neural Network with Multi-Scale Bi-directional Propagation for Video Deblurring (AAAI 2022 · similarity 0.88)
- Temporal As a Plugin: Unsupervised Video Denoising with Pre-Trained Image Denoisers (ECCV 2024 · similarity 0.87)
- Non-Local Recurrent Network for Image Restoration (NeurIPS 2018 · similarity 0.87)
- Restore From Restored: Video Restoration With Pseudo Clean Video (CVPR 2021 · similarity 0.86)
- Classic Video Denoising in a Machine Learning World: Robust, Fast, and Controllable (CVPR 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
