Home Research Papers Datasets State of the Art Pricing

Discover, visualize, and connect AI research papers. Explore the latest trends and insights in artificial intelligence research.

Product

Home
Research Papers
About

Support

Contact
Terms of Service
Privacy Policy

computer-vision

Burst Image Super-Resolution

3260 papers • 126 benchmarks • 313 datasets

Reconstruct a high-resolution image from a set of low-quality images, very like the multi-frame super-resolution task.

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in burst-image-super-resolution

Trend

Dataset

Best Model

Actions

BurstSR

SyntheticBurst

Libraries

Use these libraries to find burst-image-super-resolution models and implementations

Datasets

BurstSR

Videezy4K

Subtasks

No subtasks available.

Most implemented papers

Deep Burst Super-Resolution

L. Gool, Martin Danelljan, Goutam Bhat, Radu Timofte•Mon Jan 25 2021

A novel architecture for the burst super-resolution task, which takes multiple noisy RAW images as input, and generates a denoised, super-resolved RGB image as output by explicitly aligning deep embeddings of the input frames using pixel-wise optical flow.

152

Content

Introduction Benchmarks Datasets Subtasks Libraries Papers

Paper Graph

EBSR: Feature Enhanced Burst Super-Resolution with Deformable Alignment

Shuaicheng Liu, Haoqiang Fan, Jian Sun, Ziwei Luo, Lei Yu, Xuan Mo, Youwei Li, Lanpeng Jia, Megvii Technology•Mon May 31 2021

We propose a novel architecture to handle the problem of multi-frame super-resolution (MFSR). The proposed framework is known as Enhanced Burst Super-Resolution (EBSR), which divides the MFSR problem into three parts: alignment, fusion, and reconstruction. We propose a Feature Enhanced Pyramid Cascading and Deformable convolution (FEPCD) module to align multiple low-resolution burst images in the feature level. And then the aligned features are fused by a Cross Non-Local Fusion (CNLF) module. Finally, the SR image is reconstructed by the Long Range Concatenation Network (LRCN). In addition, we build a cascading residual pathway structure (CR) to improve the performance. We conduct several experiments to analyze and demonstrate these modules. Our EBSR model won the champion in the real track and second place in the synthetic track in the NTIRE21 Burst Super-Resolution Challenge.

70 0

Paper Graph

Unfolding the Alternating Optimization for Blind Super Resolution

Liang Wang, Zhengxiong Luo, Y. Huang, Shang Li, T. Tan•Mon Oct 05 2020

An alternating optimization algorithm, which can estimate blur kernel and restore SR image in a single model and is trained with the kernel estimated by \textit{Estimator}, instead of ground-truth kernel, thus the model could be more tolerant to the estimation error of the latter.

328 0

Paper Graph

Burst Image Restoration and Enhancement

Ming-Hsuan Yang, Syed Waqas Zamir, Salman H. Khan, Akshay Dudhane, Fahad A Khan•Wed Oct 06 2021

The central idea is to create a set of pseudo-burst features that combine complimentary information from all the input burst frames to seamlessly exchange information, and the approach initially extracts pre-processed features from each burst frame and matches them using an edge-boosting burst alignment module.

109 0

Paper Graph

Kernel-Aware Burst Blind Super-Resolution

Wenyi Lian, Shanglian Peng•Mon Dec 13 2021

A kernel-guided strategy which can solve the burst SR problem with two steps: kernel estimation and HR image restoration and a pyramid kernel-aware deformable alignment module which can effectively align the raw images with consideration of the blurry priors is introduced.

10 0

Paper Graph

BSRT: Improving Burst Super-Resolution with Swin Transformer and Flow-Guided Deformable Alignment

Shuaicheng Liu, Haoqiang Fan, Jian Sun, Ziwei Luo, Lei Yu, Youwei Li, Shen Cheng, Qi Wu, Zhihong Wen•Sun Apr 17 2022

This work combines optical flows and deformable convolutions, hence the BSRT can handle misalignment and aggregate the potential texture information in multi-frames more efficiently and wins the championship in the NTIRE2022 Burst Super-Resolution Challenge.

58 0

Paper Graph

Deep Reparametrization of Multi-Frame Super-Resolution and Denoising

L. Gool, Martin Danelljan, F. Yu, Goutam Bhat, Radu Timofte•Tue Aug 17 2021

The deep reparametrization of the maximum a posteriori formulation commonly employed in multi-frame image restoration tasks is proposed, derived by introducing a learned error metric and a latent representation of the target image, which transforms the MAP objective to a deep feature space.

73 0

Paper Graph

Towards Real-World Burst Image Super-Resolution: Benchmark and Method

Pengxu Wei, Xiang Ji, Jie Chen, Yujing Sun, Xingbei Guo, Chang Liu, Liang Lin•Fri Sep 08 2023

A large-scale real-world burst super-resolution dataset, i.e., RealBSR, is established to explore the faithful reconstruction of image details from multiple frames and a Federated Burst Affinity network (FBAnet) is introduced to investigate non-trivial pixel-wise displacements among images under real- world image degradation.

28 0

Paper Graph

Adding a benchmark result helps the community track progress.

Burst Image Super-Resolution | State-of-the-Art