Approaching a comet in 3D

As Philae approached Comet 67P/Churyumov-Gerasimenko on 12 November 2014, the lander’s downward-looking camera ROLIS took images of the descent.

Now the ROLIS team from DLR has used two of them to produce a stereographic image. To appreciate the 3D effect, the image must be viewed with red-blue/green glasses.


3D image of Comet 67P/C-G taken during the descent of Philae on 12 November 2014. Credit: ESA/Rosetta/Philae/ROLIS/DLR

This image combines two ROLIS images, acquired about an hour before the first touchdown at 15:34 GMT/16:34 CET (time onboard the spacecraft), which was confirmed on Earth at 16:03 GMT/17:03 CET. The images are separated by two minutes and the resolution is three metres per pixel.


The ROsetta Lander Imaging System (ROLIS) was developed by the DLR Institute of Planetary Research in Berlin, Germany, under the leadership of principal investigator Stefano Mottola.



  • masanori says:

    Really blown away by this one!! Take 3D glasses now if not yet! Gorgeous!! Thank you Philae!!

  • Mark G says:

    Great to have 3D images – but is there any chance of being able to download the individual pictures separately so they can be viewed through a stereoscope and not have the false colour associated with the anaglyph?

    • Jacob nielsen says:


    • Fill says:

      That’s right, individual pictures would be appreciated!

    • Bill says:

      Look 4-5 posts down, I did an X-xeyed version yesterday.

      It’s not hard to “deconstruct:” an anaglyph. Put it in an image editor (“Photoshop” or the like) and Split it into it’s individual R, G and B images, You’ll end up with three grayscale images. The Red grayscale image is the Left Eye “channel”, the Green and Blue images are identical (G&B make the cyan channel of the anaglyph) so discard one of the pair, and that Blue image makes the Right channel of the stereo pair.

      You end up with two grayscale images that can be used as crosseyed or straight-eyed stereo or in a stereoscope or stereo viewer. The advantage is that the pair is already aligned perfectly.–X-eyed-view-L.png



  • Redgy Devos says:

    I only really started to appreciate space related 3D images since the first ones from Rosetta came online.Awesome work from the teams,keep em coming!

  • Bill says:

    And here in Glorious Cross-Eyed stereo is today’s Anaglyph:–X-eyed-view-L.png

    Source: ESA/Rosetta/ROLIS
    CC: BY-SA IGO 3.0
    Stereo presentation: Bill Harris

    Whew, this is _gnarly_. Makes me happy that Little Baby Philae didn’t hop off of the Edge of the World at the bottom….


  • Roberto Perez says:

    I second Mark G, I also have a Stereoscope.

    Tip: Moving the head side to side makes the back of the comet “move”, this is obviously an illusion as the brain compensates for what it thinks is a change in perspective.

  • Hello,

    Is it possible to get the two distinct images, put them side by side, and watch them using Google Cardboard?

    Best regards,
    Sébastien Laoût.

    • Dictomail says:

      No. It’s a complicated solution to a simple problem. Their tool (I don’t mention it by name so as not to advertise it) has a complicated API and is not worth all the effort for a stereoscopic image. Their tool is geared for many images in augmented realty.

  • Adrian says:

    It is pretty amazing. I did think we may get a video in 3d. Still amazing pic.

    I need a 3d world creation program & to be able to virtually walk/float around.

  • Try without glasses as shown by NASA Goddard, You are into the picture in 3D as you know. You picture the 3D twice then from same point of view. You must move the picture in 3D taped twice to achieve 3D Cinema Thetre without Glasses. That’s all. You’ll acheve flat 3D movie clip.

  • Philaezoph says:

    Apparently not smart enough to keep the camera running for a really long time. Or have some movement trigger or anything a little bit more smart than bare stupid in the command set to be prepared for the worst.

    Such a waste.

  • Noel says:

    Very good for you!!. ESA has often been very reticent regarding public information. The “public” finances your operational costs. You are going in the right direction. The public is more educated than many may expect.. Your readership hopes to be informed in a manner close to the reality revealed by the latest data..

  • Enzo Bartolini says:


  • potty says:

    Can you post a split view version of the image so that I can see it by crossing my eyes – no 3d glasses here. Thanks!

  • Peter says:

    Yet another call for a stereo pair please!

    This is all such amazing work, and released to all like this is wonderful too. Thank you!

  • gary says:

    Has the team tried to orient rosetta’s solar panels
    at an angle with the sun to reflect light down to Philae
    to charge its batteries …

    • Robin Sherman says:

      @ Gary. Rosetta is currently about 30Km above the surface of 67P. Lets assume Philae’s location is found. Philae’s solar panels are about 1 metre square.

      Lets imagine standing on the side of a mountain with a mirror and trying to reflect the sun into a window 1 metre square on the side of a house 30Km away. Your first problem is holding the mirror accurately and steady enough to hit the target. Next because of the motion of the Earth, the angle of the Sun is continually changing. So you build a highly stable platform and build a mechanism on the top to turn the mirror at the precise speed and angle to maintain the sunbeam on the target. Possible, but it would take some serious engineering, especially if you consider the equivalent of Rosetta’s solar panels would be a mirror about 40m long.

      In this scenario, both the target and the mirror are stationary. In the case of Rosetta and Philae, both the target and the mirror are moving with respect to each other AND the Sun. OK, some serious mathematics could calculate the motions required, but Rosetta does not have the onboard processing power to continuously calculate its own position to anything like the accuracy required, nor does it have endless supplies of fuel to maintain such a flight path, let alone the incredibly fine control of its motion to achieve the accuracy required.

      If we assume all these obstacles could be overcome by Andrea and his magnificent flight control team. Then there is the issue of how much light could you actually reflect to hit the target. Firstly the solar panels are specifically designed to absorb as much sunlight as possible, so only a small percentage of the sun’s light would actually be reflected from them. This light would be unfocused and would rapidly spread out across a hemisphere whose surface area at the comet’s surface would be of the order of 5000 square Kilometres or 5 x 10^8 square metres. This means roughly 0.2 millionths of a percent of the light reflected would reach Philae’ solar panels.

      So in short, no Rosetta’s solar panels can not be used to shine light on Philae’s solar panels and charge the batteries.

  • Here, an anaglyph version with full comet (the missing slice on the right, has been virtualized):

  • Jean Hopkins says:

    Anaglyphs are great, but it would be far better to look at the 2 images, as its impossible to unscramble a anaglyph . The stereo pair has many uses especially for 3d modelling.

  • Jef says:

    with photoshop, Gimp or other image software, you can separate and record the two layers (cyan, red) and make e side by side image (or the format needed) that can be used in stereoscopic viewer.

  • Bill says:

    A How-To image on converting an anaglyph into a proper stereo pair.


    • Robin Sherman says:

      I purchased a pair of red/cyan specs for 79p. Its a lot easier.

  • Felix says:

    watching those photos by using google cardboard would be awesome!

Comments are closed.