2212 Archival Image
평가: 0+x

반응형 화상 편집기2

FROM: dempsc@foundationNet
발신: dempsc@foundationNet
TO: anathap@foundationNet
수신: anathap@foundationNet
SUBJECT: re: Image is impossible
제목: re: 불가능한 화상
TIMESTAMP: 12:07 AM, August 17, 2016
날짜: 12:07 AM, 2016년 8월 17일

Simcha (by rows, top to bottom):
심차Simcha (행, 위에서 아래로):
[1, 1, 2]
[2, 2]
[1, 1, 1]
[1, 5]
[2, 2, 2]
[2, 7]
[2, 2, 1, 2]
[1, 1, 1]
[2, 1, 1]
[2, 1, 1]

Barbara (by columns, left to right):
바바라Barbara (열, 좌에서 우로):
[2, 3, 2]
[2, 3, 2]
[1, 5]
[3, 2]
[1, 1, 1]
[3, 2, 1]
[2, 1, 1, 1]
[1, 5]

Here's the data the juniors gave me. So for Simcha's first row, the top row of the image, all I have is [1, 1, 2]. That means that, going left to right, there were possibly good superpixels, then one superpixel that had bad effects, then at least one good superpixel, then another bad one, then at least one good one, then a block of two bad ones in a row, and if there's any left in the row at that point, they're all good. We need to censor out the bad ones.
신입 애들이 준 데이터입니다. 그러니까 심차의 첫 행의 경우, 이미지 최상단의 행에 제가 가진 데이터라고는 [1, 1, 2] 뿐입니다. 그 말은, 좌에서 우로, 괜찮은 슈퍼픽셀이 좀 있을 수도 있다가, 안좋은 효과가 있는 슈퍼픽셀이 하나 있고, 하나 이상의 괜찮을 수도 있는 슈퍼픽셀이 있고, 안좋은 거 하나, 그 다음에 하나 이상의 괜찮은 거가 또 있고, 그리고는 안좋은 슈퍼픽셀 2개가 연달아 나온 뒤, 더 남은 슈퍼픽셀이 있다면 전부 괜찮은 거라는 뜻이죠. 안좋은 슈퍼픽셀들을 검열해야 합니다.

Let me know if you find anything, particularly some way to get some meaning out of what their data failed to tell us. I added a validation tool to the image edit thing that should check your censoring with their data.

See you tomorrow.

FROM: dempsc@foundationNet
TO: anathap@foundationNet
SUBJECT: Image is impossible
TIMESTAMP: 11:50 PM, August 16, 2016


Okay, I cannot get an image for this document. All our photos of 2212 made the Class Ds go crazy or something. I don't even know. Why is Dr. Ricardo even insisting on having a picture anyhow? Not like they try to have one for 96.

Anyhow, I picked the safest seeming image and had the juniors find out which bits did the insanifying. To save on subjects, they showed them increasing blocks of the picture until the delirium started up. Figure chopping it up into a 10x10 grid should be high enough resolution expunging.

Simcha went row-by-row, showing the subjects one more column with each step, until he reached the end of the row, then he went onto the next. Barbara did the same thing, but went column-by-column, adding more rows with each step. Seems reasonable? Sure, but they forgot to record where their blocks started and ended. Both of them! Where do we find these idiots?

Even so, I ought to have been able to piece together the good and bad chunks. Not so. Their measurements are ambiguous. I found maybe a dozen superpixels I can't tell if they're safe or not. And we don't have the funding to delve further (thanks, Luis). I'll send you their data. Maybe you can find something with it. If not, well, I have an adequately expunged image for the SCP that should be the answer to these damn ambiguities.

— Charlie