HDR dataset

文章目录

  • Dataset
    • NTIRE 2021 HDR (2021), CVPRW
    • HDR+ Burst Photography Dataset (2016), Google Research
  • Paper with codes
    • HDRUNet: Single Image HDR Reconstruction with Denoising and Dequantization (2021), CVPRW
    • HDR image reconstruction from a single exposure using deep CNNs (2017), ACMTG
    • A set of matlab functions for scaling of pairwise comparison experiment results based on Thurstone's model V assumptions

整理下HDR数据集 paper codes

Dataset

NTIRE 2021 HDR (2021), CVPRW

是个比赛,可以浅看一下summary。结果也给的比较清晰。

好像没有codes。

链接:https://data.vision.ee.ethz.ch/cvl/ntire21/

Built for the first challenge on high-dynamic range (HDR) imaging that was part of the New Trends in Image Restoration and Enhancement (NTIRE) workshop, held in conjunction with CVPR 2021. The challenge aims at estimating a HDR image from one or multiple respective low-dynamic range (LDR) observations, which might suffer from under- or over-exposed regions and different sources of noise.

2Tracks:

  • Single Frame HDR reconstruction
  • Multi-Frame reconstruction

HDR+ Burst Photography Dataset (2016), Google Research

发表:ACM Transactions on Graphics (Proc. SIGGRAPH Asia 2016), 35(6), 12 pp.
原始论文:https://static.googleusercontent.com/media/hdrplusdata.org/zh-CN//hdrplus.pdf
数据量:
3640 bursts,28461 images。
raw输入格式:DNG format,+ metadata
最终pipeline输出:JPG

Paper with codes

HDRUNet: Single Image HDR Reconstruction with Denoising and Dequantization (2021), CVPRW

NTIRE 2021 HDR比赛中,Track1 2nd place.

Encoder-decoder based structure, end-to-end mapping for single image HDR reconstruction, both denoising and dequantization

UNet-style based network, for using multi-scale information
condition network to perform pattern-specific modulation
a weighting network for selectively retainining information

Codes: https://github.com/chxy95/HDRUNet/blob/main/requirements.txt, PyTorch

HDR image reconstruction from a single exposure using deep CNNs (2017), ACMTG

Codes: https://github.com/gabrieleilertsen/hdrcnn, Tensorflow

A set of matlab functions for scaling of pairwise comparison experiment results based on Thurstone’s model V assumptions

Codes: https://github.com/mantiuk/pwcmp, Matlab

你可能感兴趣的:(internship,深度学习,人工智能,机器学习)