The Curiosity Super-Resolution Challenge

This is the low resolution panorama picture of the Gale Crater, the landing site of Curiosity. Here is an interesting challenge for the week-end. From the BBC interview of Mike Malin:

"The individual frames are only 144 by 144 pixels. There are 130 of them in there. It took us about an hour and six minutes to take the mosaic.

"For the full-resolution panorama, the data volume will be 64 times larger, [and] the resolution will be eight times better. But this was pretty enough and interesting enough that we thought it was worth sharing with you guys," he told BBC News.

Many of the raw images are high resolution:such as this one:

All the raw images collected so far are here.

Here is the challenge, with a bandwidth of 31MB/day, it is going to take a while to get a high resolution panorama of the Gale Crater. Can one build a high resolution panorama using some dictionary learning from the raw images and "super resolve"/inpaint the current low resolution of this panorama ?

Recent related posts of interest to this challenge:

• 90% missing pixels and you reconstructed that ?! Analysis Operator Learning and Its Application to Image Reconstruction
• Inpainting Algorithm on GitHub (TV-L2 denoising and inpainting)

Image credit: NASA/JPL-Caltech/MSSS

Liked this entry ? subscribe to Nuit Blanche's feed, there's more where that came from. You can also subscribe to Nuit Blanche by Email, explore the Big Picture in Compressive Sensing or the Matrix Factorization Jungle and join the conversations on compressive sensing, advanced matrix factorization and calibration issues on Linkedin.