Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

The Eleventh NTIRE 2026 Efficient Super-Resolution Challenge Report

Bin Ren, Hang Guo, Yan Shu, Jiaqi Ma, Ziteng Cui, Shuhong Liu, Guofeng Mei, Lei Sun, Zongwei Wu, Fahad Shahbaz Khan, Salman Khan, Radu Timofte, Yawei Li, Hongyuan Yu, Pufan Xu, Chen Wu, Long Peng, Jiaojiao Yi, Siyang Yi, Yuning Cui, Jingyuan Xia, Xing Mou, Keji He, Jinlin Wu, Zongang Gao, Sen Yang, Rui Zheng, Fengguo Li, Yecheng Lei, Wenkai Min, Jie Liu, Keye Cao, Shubham Sharma, Manish Prasad, Haobo Li, Matin Fazel, Abdelhak Bentaleb, Rui Chen, Shurui Shi, Zitao Dai, Qingliang Liu, Yang Cheng, Jing Hu, Xuan Zhang, Rui Ding, Tingyi Zhang, Hui Deng, Mengyang Wang, Fulin Liu, Jing Wei, Qian Wang, Hongying Liu, Mingyang Li, Guanglu Dong, Zheng Yang, Chao Ren, Hongbo Fang, Lingxuan Li, Lin Si, Pan Gao, Moncef Gabbouj, Watchara Ruangsang, Supavadee Aramvith

Abstract

This paper reviews the NTIRE 2026 challenge on efficient single-image super-resolution with a focus on the proposed solutions and results. The aim of this challenge is to devise a network that reduces one or several aspects, such as runtime, parameters, and FLOPs, while maintaining PSNR of around 26.90 dB on the DIV2K_LSDIR_valid dataset, and 26.99 dB on the DIV2K_LSDIR_test dataset. The challenge had 95 registered participants, and 15 teams made valid submissions. They gauge the state-of-the-art results for efficient single-image super-resolution.

The Eleventh NTIRE 2026 Efficient Super-Resolution Challenge Report

Abstract

This paper reviews the NTIRE 2026 challenge on efficient single-image super-resolution with a focus on the proposed solutions and results. The aim of this challenge is to devise a network that reduces one or several aspects, such as runtime, parameters, and FLOPs, while maintaining PSNR of around 26.90 dB on the DIV2K_LSDIR_valid dataset, and 26.99 dB on the DIV2K_LSDIR_test dataset. The challenge had 95 registered participants, and 15 teams made valid submissions. They gauge the state-of-the-art results for efficient single-image super-resolution.

Paper Structure

This paper contains 39 sections, 12 equations, 13 figures, 2 tables.

Table of Contents

  1. Introduction
  2. NTIRE 2026 Efficient Super-Resolution Challenge
  3. Dataset
  4. SPAN Baseline Model
  5. Tracks and Competition
  6. Challenge Results
  7. Main Ideas
  8. Fairness
  9. Conclusions
  10. Challenge Methods and Teams
  11. XiaomiMM
  12. Model Architecture
  13. SPABV2 Block
  14. span_attn_op: Fused CUDA Operator
  15. Near-Pixel Upsampling Branch
  16. ...and 24 more sections

Figures (13)

  • Figure 1: Team XiaomiMM: SPANV2 overall architecture. The near-pixel branch (top) provides a pixel-repeat upsampling prior, while five SPABV2 blocks (bottom) extract deep features. The two paths are concatenated (80 ch) and fused by depthwise-separable convolution before PixelShuffle$\times4$.
  • Figure 2: Team XiaomiMM: span_attn_op fuses the $1\!\times\!1$ attention convolution, element-wise addition, and element-wise multiplication into a single CUDA kernel, eliminating $3\!\times$ redundant DRAM round-trips.
  • Figure 3: Team ZenoSR: The overall architecture of Adaptive Calibration Network(ACN)
  • Figure 4: Team CUIT_HTT: Overall architecture of the proposed MambaGate-SR.
  • Figure 5: Team HAESR: The architecture of HAESR
  • ...and 8 more figures