# RealisVSR: Detail-enhanced Diffusion for Real-World 4K Video Super-Resolution

arXiv: https://arxiv.org/abs/2507.19138 · HTML: https://mlsanity.com/paper/2507.19138/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Placed in Inverse problems and image restoration, one of 14 directions in Diffusion models.
- Builds on 16 of its 27 references that are in the map.
- First posted on arXiv 25 Jul 2025.

## Placed in

- Diffusion models / [Inverse problems and image restoration](https://mlsanity.com/diffusion/inverse-problems-and-image-restoration/)

## Builds on (most cited first, 16 of 16)

- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [Scaling Rectified Flow Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2403.03206/) (2024)
- [Prompt-to-Prompt Image Editing with Cross Attention Control](https://mlsanity.com/paper/2208.01626/) (2022)
- [Frozen in Time: A Joint Video and Image Encoder for End-to-End Retrieval](https://mlsanity.com/paper/2104.00650/) (2021)
- [Real-ESRGAN: Training Real-World Blind Super-Resolution with Pure Synthetic Data](https://mlsanity.com/paper/2107.10833/) (2021)
- [Image Quality Assessment: Unifying Structure and Texture Similarity](https://mlsanity.com/paper/2004.07728/) (2020)
- [I2VGen-XL: High-Quality Image-to-Video Synthesis via Cascaded Diffusion Models](https://mlsanity.com/paper/2311.04145/) (2023)
- [OpenVid-1M: A Large-Scale High-Quality Dataset for Text-to-video Generation](https://mlsanity.com/paper/2407.02371/) (2024)
- [Exploring Video Quality Assessment on User Generated Contents from Aesthetic and Technical Perspectives](https://mlsanity.com/paper/2211.04894/) (2022)
- [Video-P2P: Video Editing with Cross-attention Control](https://mlsanity.com/paper/2303.04761/) (2023)
- [Learning Blind Video Temporal Consistency](https://mlsanity.com/paper/1808.00449/) (2018)
- [UniMax: Fairer and more Effective Language Sampling for Large-Scale Multilingual Pretraining](https://mlsanity.com/paper/2304.09151/) (2023)
- [Detail-revealing Deep Video Super-resolution](https://mlsanity.com/paper/1704.02738/) (2017)
- [DemoFusion: Democratising High-Resolution Image Generation With No ](https://mlsanity.com/paper/2311.16973/) (2023)
- [VEnhancer: Generative Space-Time Enhancement for Video Generation](https://mlsanity.com/paper/2407.07667/) (2024)
- [SeedVR: Seeding Infinity in Diffusion Transformer Towards Generic Video Restoration](https://mlsanity.com/paper/2501.01320/) (2025)

## Built on by (most cited first, 4 of 4)

- [Stream-DiffVSR: Low-Latency Streamable Video Super-Resolution via Auto-Regressive Diffusion](https://mlsanity.com/paper/2512.23709/) (2025)
- [Improved Adversarial Diffusion Compression for Real-World Video Super-Resolution](https://mlsanity.com/paper/2603.00458/) (2026)
- [LiteVSR: Lightweight Adaptation of Frozen Diffusion Transformers for Video Super-Resolution](https://mlsanity.com/paper/2606.09250/) (2026)
- [DTI: Dynamic Trajectory Initialization for Generative Face Video Super-Resolution](https://mlsanity.com/paper/2606.29198/) (2026)

## Abstract (arXiv)

Video Super-Resolution (VSR) has achieved significant progress through diffusion models, effectively addressing the over-smoothing issues inherent in GAN-based methods. Despite recent advances, three critical challenges persist in VSR community: 1) Inconsistent modeling of temporal dynamics in foundational models; 2) limited high-frequency detail recovery under complex real-world degradations; and 3) insufficient evaluation of detail enhancement and 4K super-resolution, as current methods primarily rely on 720P datasets with inadequate details. To address these challenges, we propose RealisVSR, a high-frequency detail-enhanced video diffusion model with three core innovations: 1) Consistency Preserved ControlNet (CPC) architecture integrated with the Wan2.1 video diffusion to model the smooth and complex motions and suppress artifacts; 2) High-Frequency Rectified Diffusion Loss (HR-Loss) combining wavelet decomposition and HOG feature constraints for texture restoration; 3) RealisVideo-4K, the first public 4K VSR benchmark containing 1,000 high-definition video-text pairs. Leveraging the advanced spatio-temporal guidance of Wan2.1, our method requires only 5-25% of the training data volume compared to existing approaches. Extensive experiments on VSR benchmarks (REDS, SPMCS, UDM10, YouTube-HQ, VideoLQ, RealisVideo-720P) demonstrate our superiority, particularly in ultra-high-resolution scenarios.