 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 02/12/2011 01:15 AM, Darren New wrote:
> On 12/1/2011 2:02, Invisible wrote:
>> And you're saying stuff like that actually exists in the real world, and
>> some of it is actually good quality?
>
> I've seen multiple news reports of people being PO'ed because their
> "free" photography wound up advertising something on the side of a bus.
> Perfectly legal, the way they had released it.
>
> So, yeah.
>
>>>> I can't figure out how it's even possible in theory.
>>>
>>> I had to google around for about 15 minutes before I found an actual
>>> explanation, rather than someone just saying "use the photoshop filter".
>>
>> Because, of course, everybody can afford Photoshop(r)... Oh, wait...
>
> Well, because everyone who professionally manipulates photos for video
> games can afford photoshop or some similar package.
I'm completely sure what a *professional* web designer would have access
to all sorts of things - image libraries, commercial typefaces, image
editing software, etc. My original point, way back when this thread
started, was that a guy sitting in their bedroom cannot possibly hope to
compete with or even approach this level of perfection in a finished design.
>> I still don't see how you can "fix" the seems. It's not like you can move
>> individual blades of grass around.
>
> Sure. Clone brush.
That doesn't work.
No, wait, I rephrase: I have never yet seen it work. Is that better?
>> Somebody somewhere must actually /teach/ graphic design skills...
>
> Yeah. It's called "college". :-)
...and again we're back to "I don't have that kind of money". :-/
Then again, I guess if you want a website, the thing to do is pay for a
professional to design it for you. Only trouble is, most of these guys
seem to do a fairly poor job. God only knows how you hire the good guys...
>> Then again, I spent 6 months at drawing classes, and I still can't
>> draw. So
>> maybe it's just that only a tiny fraction of the population will ever be
>> good at graphic design?
>
> Or maybe the people trying to teach you to draw were better artists than
> teachers?
Yeah, probably. Few people apparently realise that being good at
something is not the same as being good at teaching it...
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 01/12/2011 11:10 PM, Kevin Wampler wrote:
> In addition, I'm not sure what point you're making.
Everybody is like "making tilable images is trivial; you just click this
button". What I'm saying is that I cannot imagine any algorithm that the
software could be executing which would actually work.
> As an illustration of (2), I've attached a tileable image that I created
> in GIMP with about two button clicks and no manual editing. Can you spot
> the seams?
No. Which still doesn't explain how this is possible.
> As a meta-comment you have a tenancy to declare things "impossible" when
> you merely cannot see a way to achieve them. Given that your assessments
> are not infrequently wrong in this regard, you might benefit from being
> less defeatist and more curious when you don't see how something can be
> done. Just my opinion though, you're obviously free to disregard it as
> bunk.
1. Everybody has a mental model of how the world works, and based on
that model, everybody has a mostly clear idea of what is and is not
possible. If you stopped to seriously consider every single outlandish
claim hurled at you, you'd spend a long time considering utter nonsense.
Thus if somebody tells me they've solved the halting problem, that they
can remove objects from a photograph, or that they can see through
walls, I'm going to have to say that that's impossible. It's not that I
can't think of a way to do it, it's that there are strong theoretical
reasons for why it should /not/ be possible, ever.
2. If you read my original post, I mostly said that certain things "defy
comprehension". I didn't say it was "impossible". Clearly these web
designs exist - I just cannot understand *how* they can exist. I don't
see how it's possible. I was hoping that maybe somebody would be able to
explain it. Instead, everybody just said "it's all trivial, you're just
too stupid to understand that".
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 12/2/2011 1:41 AM, Invisible wrote:
>
> It's not that I
> can't think of a way to do it, it's that there are strong theoretical
> reasons for why it should /not/ be possible, ever.
Impossible in theory doesn't mean it's not often doable in practice. In
addition, I suspect you'll find that if you think carefully about your
theoretical reasons for impossibility it's not so cut and dry as you
might think, as the main problems are with the psychology of the viewer
rather than mathematical.
> 2. If you read my original post, I mostly said that certain things "defy
> comprehension". I didn't say it was "impossible".
True enough, although that changed a bit over the course of the thread.
For example just now you said "there are strong theoretical reasons
for why it should not be possible, ever". That's pretty dang close to
saying it's impossible.
> I was hoping that maybe somebody would be able to
> explain it. Instead, everybody just said "it's all trivial, you're just
> too stupid to understand that".
You misunderstand, it's not that people think you're too stupid to
figure out how to do it, it's that they think you're more than capable
of doing it but have convinced yourself that you can't and refuse to
hear otherwise -- hence the frustration. For instance, both Darren and
nemesis provided sketches of descriptions of how to do it manually,
which you promptly disregarded as well beyond your skill level. If you
sat down, watched some tutorial videos on Youtube for creating seamless
textures, and practiced with GIMP for a few days, you'd probably soon
find that you could create reasonable textures much of the time (given a
few hours of work per texture of course).
As for automated algorithms, how trivial it is depends on how
sophisticated the textures you need are. GIMP's crossfade algorithms is
totally trivial and will work well in simple cases. For more
sophisticated situations something like Adobe's content aware fill with
cyclic boundary constraints would probably do quite well (not that that
provide the source code though, so it's of little use to you).
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am 02.12.2011 10:41, schrieb Invisible:
> 1. Everybody has a mental model of how the world works, and based on
> that model, everybody has a mostly clear idea of what is and is not
> possible. If you stopped to seriously consider every single outlandish
> claim hurled at you, you'd spend a long time considering utter nonsense.
> Thus if somebody tells me they've solved the halting problem, that they
> can remove objects from a photograph, or that they can see through
> walls, I'm going to have to say that that's impossible. It's not that I
> can't think of a way to do it, it's that there are strong theoretical
> reasons for why it should /not/ be possible, ever.
As for the removal of objects from a photograph, see the sovjet russian
tradition. They were pretty good at that actually.
Not trusting anything people tell you is an important thing. But you
also need the ability to not trust your mental model of the world
either, which is a skill you don't really show in your postings.
> 2. If you read my original post, I mostly said that certain things "defy
> comprehension". I didn't say it was "impossible". Clearly these web
> designs exist - I just cannot understand *how* they can exist. I don't
> see how it's possible. I was hoping that maybe somebody would be able to
> explain it. Instead, everybody just said "it's all trivial, you're just
> too stupid to understand that".
If you want explanations, ask for them. Don't expect male people to read
that request between the lines of your posts.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Le 2011-12-01 11:32, Invisible a écrit :
>>> Interesting. As I expected, it doesn't actually produce a very
>>> convincing effect; it merely wraps the image, and then does some
>>> cross-fading. The result is a very visible transition. (Still, at
>>> least they made it circular, eh?)
>>
>> it's a smooth transition that is barely noticeable in the large scheme
>> of things. Once it's mapped on a 3D floor, you simply don't notice the
>> faded edges on the tiled floor.
>
> Maybe. But I'm talking about a 2D website background. It's pretty
> noticeable when half a pebble cross-fades into a different pebble.
>
>>> As an aside, I'd never noticed the Filter menu before. There's quite a
>>> lot of stuff in there...
>>
>> o_0
>>
>> It's just the bread and butter of bitmap editors...
>
> Well, it's the bread and butter of image processors. It won't help you
> if you're trying to (for example) remove an object from an image...
No, automatic filters can't decide what to remove on their own*. One
has to use the clone stamp tool for that (see attached).
It may come as a shock to you, but GIMP does come with tutorials and
documentation!
*Despite what they say about Photoshop CS5's or Gimp's smart removal
filter.
--
/*Francois Labreque*/#local a=x+y;#local b=x+a;#local c=a+b;#macro P(F//
/* flabreque */L)polygon{5,F,F+z,L+z,L,F pigment{rgb 9}}#end union
/* @ */{P(0,a)P(a,b)P(b,c)P(2*a,2*b)P(2*b,b+c)P(b+c,<2,3>)
/* gmail.com */}camera{orthographic location<6,1.25,-6>look_at a }
Post a reply to this message
Attachments:
Download 'clonestamp.jpg' (149 KB)
Preview of image 'clonestamp.jpg'
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> It may come as a shock to you, but GIMP does come with tutorials and
> documentation!
I spent about a week reading Grokking the GIMP. I still can't get much
more than the most basic tasks done with it. The GUI seems to have been
designed with the explicit goal of making everything maximally difficult
to understand...
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Le 2011-12-01 20:11, Darren New a écrit :
> On 12/1/2011 12:50, Orchid XP v8 wrote:
>> Yeah, I don't even want to think about what I would have to pay to get a
>> photograph of the steaming jungles of Borneo...
>
> It depends on the quality, too. Get someone who lives in Borneo to
> photograph it for you.
>
> The expensive photos are the ones you can only use once, like the photo
> you take for a package cover (like, the contents of the frozen dinner),
> especially if it takes a lot of set up (like arranging each grain of
> rice before taking the photo).
>
> You want a picture of an elephant in Africa? We have lots of those
> floating around.
>
Floating elephants? May I suggest laying off the bottle a bit?
;)
--
/*Francois Labreque*/#local a=x+y;#local b=x+a;#local c=a+b;#macro P(F//
/* flabreque */L)polygon{5,F,F+z,L+z,L,F pigment{rgb 9}}#end union
/* @ */{P(0,a)P(a,b)P(b,c)P(2*a,2*b)P(2*b,b+c)P(b+c,<2,3>)
/* gmail.com */}camera{orthographic location<6,1.25,-6>look_at a }
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 12/2/2011 8:16 AM, Francois Labreque wrote:
> No, automatic filters can't decide what to remove on their own*. One has
> to use the clone stamp tool for that (see attached).
>
> *Despite what they say about Photoshop CS5's or Gimp's smart removal
> filter.
While you are certainly correct in general, I bet Adobe's content aware
fill would have handled your example image about as well as you did
manually (which is quite well). I didn't try it though, so there's a
chance I'm wrong.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Le 2011-12-02 11:19, Invisible a écrit :
>> It may come as a shock to you, but GIMP does come with tutorials and
>> documentation!
>
> I spent about a week reading Grokking the GIMP. I still can't get much
> more than the most basic tasks done with it. The GUI seems to have been
> designed with the explicit goal of making everything maximally difficult
> to understand...
Yet, you have no problem reading mathematical proofs...
--
/*Francois Labreque*/#local a=x+y;#local b=x+a;#local c=a+b;#macro P(F//
/* flabreque */L)polygon{5,F,F+z,L+z,L,F pigment{rgb 9}}#end union
/* @ */{P(0,a)P(a,b)P(b,c)P(2*a,2*b)P(2*b,b+c)P(b+c,<2,3>)
/* gmail.com */}camera{orthographic location<6,1.25,-6>look_at a }
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 02/12/2011 04:42 PM, Francois Labreque wrote:
> Le 2011-12-02 11:19, Invisible a écrit :
>>> It may come as a shock to you, but GIMP does come with tutorials and
>>> documentation!
>>
>> I spent about a week reading Grokking the GIMP. I still can't get much
>> more than the most basic tasks done with it. The GUI seems to have been
>> designed with the explicit goal of making everything maximally difficult
>> to understand...
>
> Yet, you have no problem reading mathematical proofs...
Depends. A lot of *proofs* seem to be written in the most complicated
manner too. :-P But discussions involving mathematics? Yeah, I can
usually deal with that.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
|
|
