Author Topic: SpaceX Falcon 9 v1.1 CRS-3 Splashdown Video Repair Task Thread  (Read 1201621 times)


Offline Untribium

  • Member
  • Posts: 32
  • Switzerland
  • Liked: 32
  • Likes Given: 82
For those that aren't aware there's a new improved online editor that has all the UI for the various options and the ability to use different sources ("coerced.ts", "try1.ts") you can get to it via the wiki here http://spacexlanding.wikispaces.com/ (Online Editor v2)

If anyone has any other .ts they are using I can upload that too, just let me know (I just need the .mp4-img files and a name for it)

nice work! now I no longer have to make sure I don't accidentally modify the same block twice :D One thing though: I think it would make more sense to sort the modified blocks by y and then x as opposed to the other way around, since this is the order in which they are traversed

For those that aren't aware there's a new improved online editor that has all the UI for the various options and the ability to use different sources ("coerced.ts", "try1.ts") you can get to it via the wiki here http://spacexlanding.wikispaces.com/ (Online Editor v2)

If anyone has any other .ts they are using I can upload that too, just let me know (I just need the .mp4-img files and a name for it)

nice work! now I no longer have to make sure I don't accidentally modify the same block twice :D One thing though: I think it would make more sense to sort the modified blocks by y and then x as opposed to the other way around, since this is the order in which they are traversed

Done :) Thanks for the feedback

Offline Asmegin

  • Member
  • Posts: 37
  • Canada
  • Liked: 26
  • Likes Given: 3
For those that aren't aware there's a new improved online editor that has all the UI for the various options and the ability to use different sources ("coerced.ts", "try1.ts") you can get to it via the wiki here http://spacexlanding.wikispaces.com/ (Online Editor v2)

If anyone has any other .ts they are using I can upload that too, just let me know (I just need the .mp4-img files and a name for it)

Would some sort of toggle to highlight modified macroblocks be useful here? I find myself getting a bit confused sometimes (although my knowledge level is 1/10 so it may just be me)

Offline Chris Bergin

Looks like the news site gave this thread a bit of a boost. Gone from 67,000 to 82,000 views in 12 hours.
Support NSF via L2 -- Help improve NSF -- Site Rules/Feedback/Updates
**Not a L2 member? Whitelist this forum in your adblocker to support the site and ensure full functionality.**

Online mlindner

  • Software Engineer
  • Senior Member
  • *****
  • Posts: 2908
  • Space Capitalist
  • Silicon Valley, CA
  • Liked: 2204
  • Likes Given: 818
For those that aren't aware there's a new improved online editor that has all the UI for the various options and the ability to use different sources ("coerced.ts", "try1.ts") you can get to it via the wiki here http://spacexlanding.wikispaces.com/ (Online Editor v2)

If anyone has any other .ts they are using I can upload that too, just let me know (I just need the .mp4-img files and a name for it)

Would some sort of toggle to highlight modified macroblocks be useful here? I find myself getting a bit confused sometimes (although my knowledge level is 1/10 so it may just be me)

A mouseover pop-up for which block you are hovering over might be better (you can compare it to your list of changes anyway).
LEO is the ocean, not an island (let alone a continent). We create cruise liners to ride the oceans, not artificial islands in the middle of them. We need a physical place, which has physical resources, to make our future out there.

Offline dorkmo

  • Full Member
  • ****
  • Posts: 710
  • Liked: 338
  • Likes Given: 848
frame 5
0:0:-1,27:10:24142,17:12:30040,25:12:-1,18:13:39153,20:13:-1,25:13:35375,29:13:-1,27:15:47754,3:16:50034

added a few things from Shanuson's

Offline moralec

Could someone write a detailed instruction how to update the table with the images in the wiki http://spacexlanding.wikispaces.com/home without messing everything?

Or could someone simply put the latest image versions of:
- iframe 1 (http://forum.nasaspaceflight.com/index.php?topic=34597.msg1195314#msg1195314)
- iframe 4 (http://forum.nasaspaceflight.com/index.php?topic=34597.msg1195103#msg1195103)
- iframe 5 (http://forum.nasaspaceflight.com/index.php?topic=34597.msg1195546#msg1195546)
- iframe 6 (http://forum.nasaspaceflight.com/index.php?topic=34597.msg1195584#msg1195584 and/or http://forum.nasaspaceflight.com/index.php?topic=34597.msg1195704#msg1195704)
- iframe 8 (http://forum.nasaspaceflight.com/index.php?topic=34597.msg1195321#msg1195321 and http://forum.nasaspaceflight.com/index.php?topic=34597.msg1195464#msg1195464)
- iframe 9 (http://forum.nasaspaceflight.com/index.php?topic=34597.msg1195709#msg1195709)

Very simple thing to do.
1. click on edit on top.
2. Scroll down to the table, and find the frame you want to update.
3. Manually delete the old image
4. Add a new image (there is a button for that).  There is no need to upload anything, just make a reference to the image that is stored here in nsf.
5. Add the name of the author and new code.
6. Click save.

Offline moralec

frame 5
0:0:-1,27:10:24142,17:12:30040,25:12:-1,18:13:39153,20:13:-1,25:13:35375,29:13:-1,27:15:47754,3:16:50034

added a few things from Shanuson's

Very nice. Don't forget to add it to the wiki!

Offline CraigLieb

  • Full Member
  • ****
  • Posts: 1193
  • Dallas Fort Worth
  • Liked: 1349
  • Likes Given: 2394
Hi guys,

This is just for inspiration:

"We've always defined ourselves by the ability to overcome the impossible"

Regards,

arnezami

Terrific video.

And also useful to note the side view of the rocket descending into the smoke,
 much like the stage returning as it hits the surface of the water.

I suppose this is obvious to the rocket scientists that we would see just parts of the struts.

The only problem is that in iframe 13, you can see both being the same size again....


Weird....

http://spacexlanding.wikispaces.com/

Waves pass,  how much time between iframes?
On the ground floor of the National Space Foundation... Colonize Mars!

Offline moralec

Hi guys,

This is just for inspiration:

"We've always defined ourselves by the ability to overcome the impossible"

Regards,

arnezami

Terrific video.

And also useful to note the side view of the rocket descending into the smoke,
 much like the stage returning as it hits the surface of the water.

I suppose this is obvious to the rocket scientists that we would see just parts of the struts.

The only problem is that in iframe 13, you can see both being the same size again....


Weird....

http://spacexlanding.wikispaces.com/

Waves pass,  how much time between iframes?

About one second

Offline watermod

  • Full Member
  • ****
  • Posts: 519
  • Liked: 177
  • Likes Given: 153
I haven't done much with mpeg-4 but have done work with other forms of imaging (3D).   To me, one fact that sticks in the mind is that the actual upper part of the rocket should be the same (excepting albedo) from the beginning of the flight  until the end.  That should be able to provide clues or hints to reconstructing the images.  I realize that being a 2D and not a 3D image the albedo is not as easy to factor out of the image.  It would be interesting to have a 3D image of the first stage, layer the good frames early in flight  upon it's surface as an albedo cutting out the sky, ocean and such.  Then factor that back into the MPEG-4 images (both as direct missing bits and sunlight and sun angle hints).  Other then bits like the legs as they disappear into the ocean everything else should be the background sky and direct light on the first stage that is different from the light at launch.  Same for the static dirt on the lens.   It's a hint.  They are all hints to the missing bits.  From those missing bits you might get enough hints to reconstruct the ocean and remaining scenery.
(example of albedo being manipulated as a 3D skin to recover useful information: http://www.engr.uky.edu/~lgh/data/dataGraveStones2010.htm)





Offline seanpg71

  • Member
  • Posts: 38
  • Liked: 49
  • Likes Given: 22
frame 5
0:0:-1,27:10:24142,17:12:30040,25:12:-1,18:13:39153,20:13:-1,25:13:35375,29:13:-1,27:15:47754,3:16:50034

added a few things from Shanuson's


That's helpful and should save me a bit of time.  I've been working from the bottom up on 5.  It's a bit of a mess.  Mostly I've just been trying to fix the bit alignment issues so that a picture shows up so far.  This is as far as I've gotten thus far.

0:0:-1,40:17:56308,43:19:69441,0:21:106970,1:21:76873,11:22:85251,12:22:85489,26:22:86772,
27:22:87056,37:22:87982,38:22:88152,10:23:90705,11:23:91098,16:23:91462,17:23:91682,
22:24:97633,23:24:97869,0:25:100325,0:27:106003,2:27:105388,22:27:110276,30:28:121538,
31:28:121730,1:29:123050,2:29:123876


Offline Asmegin

  • Member
  • Posts: 37
  • Canada
  • Liked: 26
  • Likes Given: 3
frame 5
0:0:-1,27:10:24142,17:12:30040,25:12:-1,18:13:39153,20:13:-1,25:13:35375,29:13:-1,27:15:47754,3:16:50034

added a few things from Shanuson's


That's helpful and should save me a bit of time.  I've been working from the bottom up on 5.  It's a bit of a mess.  Mostly I've just been trying to fix the bit alignment issues so that a picture shows up so far.  This is as far as I've gotten thus far.

0:0:-1,40:17:56308,43:19:69441,0:21:106970,1:21:76873,11:22:85251,12:22:85489,26:22:86772,
27:22:87056,37:22:87982,38:22:88152,10:23:90705,11:23:91098,16:23:91462,17:23:91682,
22:24:97633,23:24:97869,0:25:100325,0:27:106003,2:27:105388,22:27:110276,30:28:121538,
31:28:121730,1:29:123050,2:29:123876

Great start. Added to Wiki

Offline dorkmo

  • Full Member
  • ****
  • Posts: 710
  • Liked: 338
  • Likes Given: 848
frame 5
0:0:-1,27:10:24142,17:12:30040,25:12:-1,18:13:39153,20:13:-1,25:13:35375,29:13:-1,27:15:47754,3:16:50034

added a few things from Shanuson's


That's helpful and should save me a bit of time.  I've been working from the bottom up on 5.  It's a bit of a mess.  Mostly I've just been trying to fix the bit alignment issues so that a picture shows up so far.  This is as far as I've gotten thus far.

0:0:-1,40:17:56308,43:19:69441,0:21:106970,1:21:76873,11:22:85251,12:22:85489,26:22:86772,
27:22:87056,37:22:87982,38:22:88152,10:23:90705,11:23:91098,16:23:91462,17:23:91682,
22:24:97633,23:24:97869,0:25:100325,0:27:106003,2:27:105388,22:27:110276,30:28:121538,
31:28:121730,1:29:123050,2:29:123876

heres my most recent

0:0:-1,27:10:24142,16:12:30040,24:12:-1,17:13:39153, 20:13:-1,25:13:35375,29:13:-1,28:16:47754,0:17:50034, 43:19:-1,0:26:103820,13:27:-1,17:27:109631,25:27:-1, 27:27:110752,28:28:-1,33:28:121771,2:29:124970
« Last Edit: 05/09/2014 04:45 am by Lar »

Offline Asmegin

  • Member
  • Posts: 37
  • Canada
  • Liked: 26
  • Likes Given: 3
Is anyone working on 3 or 4? Those are the two bad ones. 5 was bad but the last few posts show progress!

I searched Frame 3 for a while but couldn't find anything. Unrecoverable?

Offline dorkmo

  • Full Member
  • ****
  • Posts: 710
  • Liked: 338
  • Likes Given: 848
Is anyone working on 3 or 4? Those are the two bad ones. 5 was bad but the last few posts show progress!

I searched Frame 3 for a while but couldn't find anything. Unrecoverable?

judging from whats become of frame 5, i think frame 4 is going to be the most important for getting a nice shot of the leg deployment. and im afraid 3 is a lost cause. :(

i noticed that some are using a different ts file, try1? are we going to be in a world of hurt if we have mixed data or will it be okay?

ps this online editor on the wiki is SPECTACULAR! i wouldnt have been able to figure out doing it the hard way. kudos!


--edit--
added 22:24:97869 and changed some around the bottom:

0:0:-1,27:10:24142,16:12:30040,24:12:-1,17:13:39153,20:13:-1, 25:13:35375,29:13:-1,28:16:47754,0:17:50034,43:19:-1, 22:24:97869,02:27:-1,24:27:110437,13:27:-1,17:27:109631, 25:27:-1,27:27:110752,28:28:-1,33:28:121771,2:29:124970

Edit/Lar: please put spaces in every once in a while so your lines wrap. I think the parms are space agnostic (if they are after the comma but not anywhere else anyway?). thanks
« Last Edit: 05/09/2014 04:43 am by Lar »

Offline grythumn

  • Full Member
  • **
  • Posts: 243
  • Liked: 167
  • Likes Given: 246
Minor tweak to iframe13, try1: X:90932:80,40:1:-1,0:10:-1:0:-10,1:10:-1:,4:10:49027,8:29:-1

Offline dorkmo

  • Full Member
  • ****
  • Posts: 710
  • Liked: 338
  • Likes Given: 848
Is anyone working on 3 or 4? Those are the two bad ones. 5 was bad but the last few posts show progress!

I searched Frame 3 for a while but couldn't find anything. Unrecoverable?

frame 4, so far only found a little bit of the left side and maybe a few blocks of the right
00:14:-1,05:01:-1,35:05:-1,19:05:20944,35:05:-1,43:06:27582,0:10:-1,21:13:50333,25:14:57756 ,08:15:61126,36:15:-1,00:17:71044,0:24:-1,00:27:114198


Offline theshadow27

  • Member
  • Posts: 28
  • Liked: 27
  • Likes Given: 7
Wow just read through the 27 pages and am very impressed.

Trying to work on iframe 4, as a scam guy, I was thinking of trying to develop a brute-force script to match the 19:35 timestamp. From what I have (quickly) counted, the first four blocks in the bottom row have known data. I don't have mpeg experience, but from what I've gathered from this thread, due to the down-right cascade this can only directly check the first four blocks in each row. AFAIU, corrections in those blocks could potentially help resolve the entire frame.

From the 704x480 PNG output of the custom ffmpeg command (using MMB on frames 6, 11, 12, and 13), the known pixel counts are:

0: 36/576
1: 148/576
2: 130/576
3: 154/576

Total: 468/2304 = ~.203

or about 20% which is huge in terms of a known-plaintext attack. Add in that ffmpeg errors should probably decrease with a valid flip, I think a brute force is computationally feasible. A tremendous speedup would be possible with a heuristic search algorithm, which I assume those of you that have been playing manually probably have a mental version of.

Anyone care to take a swing at describing your process of coming up with the next MMB token given the existing tokens and the next error?




Tags:
 

Advertisement NovaTech
Advertisement Northrop Grumman
Advertisement
Advertisement Margaritaville Beach Resort South Padre Island
Advertisement Brady Kenniston
Advertisement NextSpaceflight
Advertisement Nathan Barker Photography
0