Project Page: Imaging With Nature

Photograph 1: Set-up used in the experiment

Imaging With Nature: A Universal Analog Compressive Imager Using a Multiply Scattering Medium

Antoine Liutkus, David Martina, Sébastien Popoff, Gilles Chardon, Ori Katz, Geoffroy Lerosey, Sylvain Gigan, Laurent Daudet, Igor Carron

Table of content:
  1. The Summary
  2. The Paper
  3. The Preprint
  4. The Data / Photos and Videos
  5. The Discussions and Post Publication Peer Reviews
  6. The Story Behind this Publication
  7. The Press Releases  / Coverage

1. The Summary

Compressive sensing is a mathematical framework that allows the acquisition and recovery of the sparsest solution to underdetermined systems of linear equations through the use of non-combinatorial solvers. Ever since the first papers on the subject (10 years ago), there have been discussions on the type of "measurement matrices" that are acceptable for this task. Very early on, random matrices were seen as mathematically sound "measurement matrices" and this led to several implementations using very well engineered systems mimicking random gaussian matrices. A famous example is the use of DMDs to produce the randomness in the single pixel camera at Rice University.  In this paper, we use the natural scattering of light when it goes through an opaque material to perform a similar mixing. In effect, since one snapshot results in obtaining several combination of light rays, our operation is "embarassingly parallel" compared to the sequential mode of acquisition of most other compressive systems.
In the paper, we also showed a map made of experimental datapoints showing when and if information that is transferred through this opaque material can yield back the original image. We believe this is a first. Recent theoretical/empirical phase transition curves have been found to overlap our experimental findings

2. The Paper:

Published on Scientific Reports as Open Access:

Scientific Reports 4, Article number: 5552 doi:10.1038/srep05552
Received 28 January 2014 
Accepted 13 June 2014 
Published 09 July 2014

3. The Preprint

Imaging With Nature: A Universal Analog Compressive Imager Using a Multiply Scattering Medium

(Submitted on 2 Sep 2013)

The recent theory of compressive sensing leverages upon the structure of signals to acquire them with potentially much fewer samples than was previously thought necessary. However, most current sensing hardware developed to take advantage of this framework for imaging revolve around carefully engineered sequential sampling procedures aimed at producing randomness. Instead, we propose the use of the naturally occurring randomness of wave propagation in multiply scattering media as an optimal compressive imaging mechanism. Waves reflected from an object are detected after propagation through a well-characterized complex medium. Each local measurement thus contains global information about the object, yielding a purely analog compressive sensing method. We experimentally demonstrate the effectiveness of the proposed approach for optical imaging by using a 300-micrometer thick layer of white paint as the compressive imaging device. Scattering media are thus promising candidates for designing efficient and compact compressive imagers.

4. The Data / Photos and Videos

The data will be released in the near future.

Photograph 2: Elements annotated in the experiment

Compressive Sensing Using Speckle

Video 1:
 The following video is made up of a series of frames recorded on the focal plane array of the camera. For each frame, the image represents the same object in front of the multiple scattering medium. One can see that the frame change over time in a vibration like manner. This is an early recording.

5. The Discussions and Post Publication Peer Reviews

There are multiples ways to provide a view on this work:
  1. Comment at PubMedCommons
  2. PubPeer:
  3. Be a Peer reviewer on
  4. Comment on the compressive sensing subreddit ( one of the author - IC - "controls" the content of this site )
  5. Comment on the Google+ group ( one of the author - IC - "controls" the content of this site )
  6. Comment in the comment section of this blog entry ( one of the author - IC - "controls" the content of this site )
If you want to remain anonymous and provide an input on this publication, use 1 through 3 or 4 (with a throwaway account) or 5 with an anonymous name. There is the possibility that in solution 4 through 6, your input might get deleted because of 

Related discussions on Nuit Blanche (one of the co-author, Igor Carron, is editor of Nuit Blanche )

6. The Story Behind this Publication

upcoming. Grab some popcorn in advance !

7. The Press Releases/ Coverage

Coverage on Nuit Blanche

web analytics