StereoMSI

Collection Title:

StereoMSI

Collection Description:

1. Multi-spectral super-resolution dataset (extension to PIRM2018 multi-spectral super-resolution challenge dataset). The dataset includes - 272 training stereo registered multi-spectral RGB image pairs including the original RGB mosaic images. - 970 raw multi-spectral images. These need to be downsampled (x2 or x3 etc.) to create low-resolution im... moreages for training a CNN for the task of super-resolution. - 50 test images. 2. Dataset for simultaneous colour-prediction and super-resolution. This dataset includes: - 250 registered stereo multi-spectral/RGB image pairs for training - 25 registered stereo multi-spectral/RGB image pairs for validation - 21 registered stereo multi-spectral/RGB image pairs for testing. less

Field of Research:

Computer Vision ;  Image Processing ;  Knowledge Representation and Machine Learning

DOI:

https://doi.org/10.25919/5e17c1f7928d7

Start Date:

2017

End Date:

2018

Contact:CSIRO Enquiries
 CSIROEnquiries@csiro.au
 1300 363 400
Keywords:

Super-resolution ;  SR ;  Multi-spectral ;  RGB

Related Links:

:	Super-resolved Chromatic Mapping of Snapshot Mosaic Image Sensors via a Texture Sensitive Residual Network
:	PIRM2018 Challenge on Spectral Image Super-Resolution: Dataset and Study

Supporting Files:

Shoeiby_PIRM2018_Challenge_on_Spectral_Image_Super-Resolution_Dataset_and_Study_ECCVW_2018_paper.pdf

 

03570154_wavelegnth_in_block

Lineage:

1. Images in the multi-spectral super-resolution dataset are in .tiff format. 2. Images in the colour-prediction dataset are in .tiff (RGB) and .mat (multi-spectral) formats. Converting .mat to .npy can accelerate training. See https://arxiv.org/pdf/1909.02221.pdf for more information on multi-spectral data format.

Licence:

CSIRO Data Licence

Organisations:

CSIRO (Australia)

Attribution Statement:

Shoeiby, Mehrdad (2019): StereoMSI. v2. CSIRO. Data Collection. https://doi.org/10.25919/5e17c1f7928d7

Rights Statement:

All Rights (including copyright) CSIRO 2019.

Access:

The metadata and files (if any) are available to the public.