Research map / Video / Video segmentation and tracking
Investigating Tradeoffs in Real-World Video Super-Resolution
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 image pre-cleaning before temporal propagation | BasicVSR for real-world VSR · VideoLQ | NIQE | Helped |
| Remove the image cleaning loss (ablation) | RealBasicVSR · Real-world VSR | output quality | Hurt |
| Replace the image-based cleaning module with a recurrent network (ablation) | RealBasicVSR · Real-world VSR | output quality | Hurt |
| Use dynamic refinement instead of fixed cleaning iterations (ablation) | RealBasicVSR cleaning module · VideoLQ | NIQE | Helped |
| Use stochastic degradation instead of the conventional training scheme | RealBasicVSR training · Stochastic degradation comparison | NIQE: 4.7191 -> 4.6836 | Helped |
| Use stochastic degradation instead of the conventional training scheme | RealBasicVSR training · Stochastic degradation comparison | Time per iteration: ~2.5s -> ~1.5s | Helped |
| Flip the sequence without adding temporally varying degradations (ablation) | RealBasicVSR training · Stochastic degradation comparison | NIQE: 4.7191 -> 4.6926 | Helped |
| Train with longer sequences rather than larger batches (ablation) | RealBasicVSR · VideoLQ | visual quality | Helped |
| Add video compression to the training degradations (ablation) | RealBasicVSR · Real-world VSR | performance | Helped |
Closest papers by meaning
- NegVSR: Augmenting Negatives for Generalized Noise Modeling in Real-world Video Super-Resolution (AAAI 2024 · similarity 0.92)
- DOVE: Efficient One-Step Diffusion Model for Real-World Video Super-Resolution (NeurIPS 2025 · similarity 0.90)
- Towards Realistic Data Generation for Real-World Super-Resolution (ICLR 2025 · similarity 0.89)
- One-Step Diffusion for Detail-Rich and Temporally Consistent Video Super-Resolution (NeurIPS 2025 · similarity 0.89)
- Unsupervised Diffusion-Based Degradation Modeling for Real-World Super-Resolution (AAAI 2025 · similarity 0.89)
- SEAL: A Framework for Systematic Evaluation of Real-World Super-Resolution (ICLR 2024 · similarity 0.89)
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.