AI Paper Atlas

Research map / Benchmarks / UDM10

UDM10 leaderboard

14 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on UDM10. Best LPIPS: 0.13 (WEVSR, ICML 2026).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

UDM10 · LPIPS

13 papers, each with its best reported number for its own method; lower is better.

#LPIPS ↓MethodPaperTopic
10.13WEVSR
WAN-2.1
WEVSR: Video Diffusion Generators for Real-World Video Super‑Resolution with Wavelet-Enhanced VAE Encoder
ICML 2026 · Table 1
Super-resolution
20.133Upscale-A-Video
SD ×4 Upscaler
Upscale-A-Video: Temporal-Consistent Diffusion Model for Real-World Video Super-Resolution
CVPR 2024 · Table 1
Video generation
30.1885STAR
CogVideoX-5B
STAR: Spatial-Temporal Augmentation with Text-to-Video Models for Real-World Video Super-Resolution
ICCV 2025 · Table 1
Video segmentation and tracking
40.2156SCST
Stable Diffusion V2.1
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
CVPR 2025 · Table 1
Video segmentation and tracking
50.242ART-VSR
Wan2.1
ART-VSR: Adaptive Rectified Trajectories for One-Step Video Super-Resolution
ECCV 2026 · Table 1
Video segmentation and tracking
60.2444PS-SRPS-SR: Pseudo-Single-Step Video Super-Resolution via Speculative Diffusion
CVPR 2026 · Table 1
Super-resolution
70.2678TRaM-VSR (Ours)TRaM-VSR: Importance-Aware Token Routing and Merging for One-Step Diffusion Video Super-Resolution
ECCV 2026 · Table 1
Video segmentation and tracking
80.2696S1+S2-I/V
CogVideoX1.5
DOVE: Efficient One-Step Diffusion Model for Real-World Video Super-Resolution
NeurIPS 2025 · Ablation table, training strat
Super-resolution
90.271DTG-Restore (Ours)DTG-Restore: Training-Free Diffusion Refinement for Generative Video Super-Resolution
CVPR 2026 · Table 1
Video generation
100.2972InfVSR
Wan 2.1 1.3B
InfVSR: Toward Consistency-Driven Streaming Generative Video Super-Resolution
ICML 2026 · Table 1
Video generation
3 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

UDM10 · PSNR

11 papers, each with its best reported number for its own method; higher is better.

#PSNR ↑MethodPaperTopic
130.79Upscale-A-Video
SD ×4 Upscaler
Upscale-A-Video: Temporal-Consistent Diffusion Model for Real-World Video Super-Resolution
CVPR 2024 · Table 1
Video generation
228.1WEVSR
WAN-2.1
WEVSR: Video Diffusion Generators for Real-World Video Super‑Resolution with Wavelet-Enhanced VAE Encoder
ICML 2026 · Table 1
Super-resolution
327.2HQ-VSR
CogVideoX1.5
DOVE: Efficient One-Step Diffusion Model for Real-World Video Super-Resolution
NeurIPS 2025 · Ablation table, training datas
Super-resolution
426.73TRaM-VSR (Ours)TRaM-VSR: Importance-Aware Token Routing and Merging for One-Step Diffusion Video Super-Resolution
ECCV 2026 · Table 1
Video segmentation and tracking
526.62SparkVSR (no ref)
CogVideoX1.5-5B I2V
SparkVSR: Interactive Video Super-Resolution via Sparse Keyframe Propagation
ECCV 2026 · Table 1
Video segmentation and tracking
626.42SCST
Stable Diffusion V2.1
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
CVPR 2025 · Table 1
Video segmentation and tracking
726.42RealViformer (Ours)RealViformer: Investigating Attention for Real-World Video Super-Resolution
ECCV 2024 · Table 1
Video segmentation and tracking
825.61DTG-Restore (Ours)DTG-Restore: Training-Free Diffusion Refinement for Generative Video Super-Resolution
CVPR 2026 · Table 1
Video generation
924.86InfVSR
Wan 2.1 1.3B
InfVSR: Toward Consistency-Driven Streaming Generative Video Super-Resolution
ICML 2026 · Table 1
Video generation
1023.913PS-SRPS-SR: Pseudo-Single-Step Video Super-Resolution via Speculative Diffusion
CVPR 2026 · Table 1
Super-resolution
1 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.