live2d can animate a picture and make small movements, just as they show in this video.
1 Does anyone know how it works?
2 Is there any paper describe the mechanism behind it? I tried google scholar search but find few.
3 Is there any open source work on this field?
The fundamental algorithm is like control points in Adobe Illustrator. Control points are like anchors for the image. You can shrink, stretch and bend the image by moving the control points.
Unfortunately, NO. Live2D is all developed in the company.
For now it is all closed project.
Related
I want to remove the regular strips of the image as shown as follow. I try many methods, and they do not work, such as image media filter and FFT filter.
Could you tell me how to remove the strips?
All that black is removing a ton of information from the image. You have two options available - either re-capture that missing information in a new shot, or attempt to invent / synthesize / extrapolate the missing information with software.
If you can re-shoot, get your camera as close to the mesh fence as you can, use the largest aperture your lens supports to have the shallowest possible depth of field, and set your focus point as deep as possible - this will minimize the appearance of the mesh.
If that is the only still you have to work with, you've got a few dozen hours of playing with the clone and blur tools in front of you in just about any image editing software package you like.
Photoshop would be my go to tool of choice for this. In Photoshop CS5 they introduced something called content aware fill. I'm not sure if it will help you in this specific case because there is SO MUCH black that Adobe's algorithm may think other parts of the mesh are valid sources for filling in the mesh you're trying to clear out.
I need to start building a image application and my customer wants to arrange the picutes in the screen like google tv does, as well as everpix. I have been looking for it for a while but I was unable to find it. The result of arranging the pictures this way is amazing and makes the best use of the screen space.
http://www.google.com//tv/static/images/photos_tv_straight.png
Is this a known algorithm? Does it have a name?
Many thanks
T
Like jwpat7 suggested look for "photo collage layout" algorithms. Particularly things like "treemap" and similar (squarified trieemap). I am working on similar algorithm and for some small number of images you just need to solve simple system of linear equations. There is another HP article that is probably more close to what are you looking for.
Mixed-Initiative Photo Collage Authoring - look at part 4.
Following image is done with some squarified treemap and ratio optimization.
Search for photo montage and photo collage algorithms, as well as photo tiling.
An HP article called "Structured Layout for Resizable Background Art" may be helpful.
Numerous collage programs are available for purchase and some software is available in source form; e.g. see hlrnet list, software.informer list, and perhaps this resizing blurb.
The algebra for scaling photos for a collage while maintaining aspect ratios is straightforward and easily described for specific cases, but not for too-general ones.
In css you can arrange images from horizontal to vertical. A good example is the Google image search. There is the Jquery Masonry plugin to arrange from vertical to horizontal and it has some nice animation. In your example you want to have rather a rectangle arrangement I suggest a treemap algorithm where you can also rotate the rectangle in 90°.
I'm sorry if my question is somewhat vague. It's been a few years since I did anything with Qt, and back then I never did any fancy image stuff. What I'm asking for below is just some general suggestions on which classes to consider using. I'm trying to avoid barking up the wrong tree from the very start.
The situation: I'm writing a Qt-based program in which I need to display a somewhat large (let's say 5000x5000) raster image. The user should be able to zoom (quickly) in and out, and pan around the image in a way similar to for example Google maps. So far, this is not very different from the Qt ImageViewer example, except perhaps for the requirement that zooming happens quickly. However, I need to draw on the order of 50k simple geometric shapes (let's say circles) on top of the image, and be able to add and remove some of these in a simple way. The circles should have the same size no matter the zoom level, and should thus either be redrawn whenever the user zooms, or should be drawn with vector graphics. Think of the circles as map annotations. These should look the same at any zoom level, and also behave nicely with respect to panning.
I guess my question is twofold:
Can Qt draw vector graphics on top of a raster image?
In general, which classes should I consider for the above?
Thanks in advance. I don't like answering vague questions myself, but maybe someone with experience with Qt's graphics capabilities has an answer.
I suggest you use QGraphicsView and friends for this. It helps handling all the view/world transformation and the vector items can be achieved with various QGraphicsItems.
You can change the sizes of the items whenever the zoom level changes to maintain constant apparent sizes.
In the upcoming version of Photoshop there is a feature called Content-Aware fill.
This feature will fill a selection of an image based on the surrounding image - to the point it can generate bushes and clouds while being seamless with the surrounding image.
See http://www.youtube.com/watch?v=NH0aEp1oDOI for a preview of the Photoshop feature I'm talking about.
My question is:
How does this feature work algorithmically?
I am a co-author of the PatchMatch paper previously mentioned here, and I led the development of the original Content-Aware Fill feature in Photoshop, along with Ivan Cavero Belaunde and Eli Shechtman in the Creative Technologies Lab, and Jeff Chien on the Photoshop team.
Photoshop's Content-Aware Fill uses a highly optimized, multithreaded variation of the algorithm described in the PatchMatch paper, and an older method called "SpaceTime Video Completion." Both papers are cited on the following technology page for this feature:
http://www.adobe.com/technology/projects/content-aware-fill.html
You can find out more about us on the Adobe Research web pages.
I'm guessing that for the smaller holes they are grabbing similarly textured patches surrounding the area to fill it in. This is described in a paper entitled "PatchMatch: A Randomized Correspondence Algorithm for Structural Image Editing" by Connelly Barnes and others in SIGGRAPH 2009. For larger holes they can exploit a large database of pictures with similar global statistics or texture, as describe in "Scene Completion Using Millions of Photographs". If they somehow could fused the two together I think it should work like in the video.
There is very similar algorithm for GIMP for a quite long time. It is called resynthesizer and probably you should be able to find a source for it (maybe at the project site)
EDIT
There is also source available at the ubuntu repository
And here you can see processing the same images with GIMP: http://www.youtube.com/watch?v=0AoobQQBeVc&feature=related
Well, they are not going to tell for the obvious reasons. The general name for the technique is "inpainting", you can look this up.
Specifically, if you look at what Criminisi did while in Microsoft http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.67.9407 and what Todor Georgiev does now at Adobe http://www.tgeorgiev.net/Inpainting.html, you'll be able to make a very good guess. A 90% guess, I'd say, which should be good enough.
I work on a similar problem. From what i read they use "PatchMatch" or "non-parametric patch sampling" in general.
PatchMatch: A Randomized Correspondence Algorithm
for Structural Image Editing
As a guess (and that's all that it would be) I'd expect that it does some frequency analysis (some like a Fourier transform) of the image. By looking only at the image at the edge of the selection and ignoring the middle, it could then extrapolate back into the middle. If the designers choose the correct color plains and what not, they should be able to generate a texture that seamlessly blends into the image at the edges.
edit: looking at the last example in the video; if you look at the top of the original image on either edge you see that the selection line runs right down a "gap" in the clouds and that right in the middle there is a "bump". These are the kind of artifacts I'd expect to see if my guess is correct. (OTOH, I'd also expect to see them is it was using some kind of sudo-mirroring across the selection boundary.)
The general approach is either content-aware fill or seam-carving. Ariel Shamir's group is responsible for the seminal work here, which was presented in SIGGRAPH 2007. See:
http://www.faculty.idc.ac.il/arik/site/subject-seam-carve.asp
Edit: Please see answer from the co-author of Content-Aware fill. I will be deleting this soon.
I know this is leaning more in the direction of a designer question, but as I am faced with developing something which requires me to crop an image, I thought I would give the question a shot.
This seems like a ridiculous question to ask, but I've look all over the IDE (Expression Blend 2) to try and find a way to crop my image, but I can't figure it out.
This seems to be very much in line with Joel's question and is discussed in Podcast 58 in the sense that I'm a complete noob when it comes to designing in Expression Blend. I am adamantly interested in figuring out the most efficient way to do this. I found an article that describes a work flow you can go through that will produce a crop, which I added as an answer below, but I'm really hoping someone else will know of a quicker (less clicks) way to do something as trivial as this.
Does anyone know how this can be done?
As far as I know, there's no way to crop an image directly in expression Blend. Blend is not an image editing application. You need another tool for that.
What you can do though, is clipping an image if you only want to show a portion of it. Just add a rectangle on top of it right-click it, go to path -> make clipping path.
alt text http://img200.imageshack.us/img200/7370/example1.jpg
Now select the System.Windows.Controls.Image entry from the list you want to apply the clipping on and hit Ok
You can even use rounded rectangles, circles and custom paths to clip, but in most cases a rectangle will do the trick.
Just ran into another way.
Have a look at this question. It uses a CroppedBitmapClass as the source of an image. It's not actual drawing in Blend, but you can add it by hand editing Xaml. From your question it's not clear if you are creating a Silverlight or a WPF application in Expression Blend. The CroppedBitmapClass is available in WPF only.
With the new Silverlight 3 you can use the WritableBitmap to do image cropping:
I found an article that has steps to do an image crop, but it's very drawn out to do such a simple operation. You would think something that MS Paint can do in a couple button clicks would be similarly easy in Blend.
Here's the link.
I'm still wondering if there's an easier way to do this, however.
The other problem with this approach is that afterwards, I can't change the size of the rectangle that I'm cropping the image with, which I need to be able to do, because I have to have the image be an exact number of pixels in width and height.