NVIDIA@I3D：SRAA_子像素重构抗锯齿技术白皮书、文档、视频出笼

Edison

Subpixel Reconstruction Antialiasing

http://research.nvidia.com/publi ... uction-antialiasing

"Subpixel Reconstruction Antialiasing"
Matthäus G. Chajdas (Technische Universität München and NVIDIA), Morgan McGuire (NVIDIA), David Luebke (NVIDIA), in i3D, February 2011
Research Area:         3D Graphics
Author(s):         Matthäus G. Chajdas (Technische Universität München and NVIDIA), Morgan McGuire (NVIDIA), David Luebke (NVIDIA)
Date:         February 2011
Download(s):         
        

Abstract:         

Subpixel Reconstruction Antialiasing (SRAA) combines single-pixel (1x) shading with subpixel visibility to create antialiased images without increasing the shading cost.

SRAA targets deferred-shading renderers, which cannot use multisample antialiasing.

SRAA operates as a post-process on a rendered image with superresolution depth and normal buffers, so it can be incorporated into an existing renderer without modifying the shaders.

In this way SRAA resembles Morphological Antialiasing (MLAA), but the new algorithm can better respect geometric boundaries and has fixed runtime independent of scene and image complexity. SRAA benefits shading-bound applications.

For example, our implementation evaluates SRAA in 1.8 ms (1280x720) to yield antialiasing quality comparable to 4-16x shading. Thus SRAA would produce a net speedup over supersampling for applications that spend 1 ms or more on shading; for comparison, most modern games spend 5-10 ms shading. We also describe simplifications that increase performance by reducing quality.

huanghuooo01

看不懂，求翻译帝
难道N卡要支持新的抗锯齿技术

Edison

简单地说，就是 SRAA 类似于 MLAA，但是使用了新的算法可以更多地顾及几何体界线并且执行时间无视场景独立性和画面复杂度是固定的，如果游戏是 shader 饥渴类的程序，SRAA 的执行成本将会相对较低。

SRAA 可以在 1280x720 分辨率下以 1.8 毫秒每帧的速率完成，可以达到相当于 4-16 倍渲染分辨率的效果。

lysine1

只關心SRAA會影響字體麽，MLAA會讓字體都發園這點太惡，眼睛實在受不了。

disruptor

就是有 1.8 毫秒的延时吗

goldman948

难道也是糨糊aa?

scim

新算法频频出现啊

BDFMK2

MLAA的算法本来就可以改进的吧，只不过换个名字而已

SnakeLee

现在看，NV的后发“我们也有”系列都带点打脸属性，相对其竞争者而言

3118595

抵制一切浆糊AA   不管是哪家的   当然如果实际使用好的话热烈欢迎-_-

那啥  这对显卡要求小点么？

66666

SRAA根CSAA比较有何优势呢？

disruptor

回复 66666 的帖子

速度快，基本无需什么显卡资源，属于一种后处理技术

yyzjp

昨天晚上做了个奇怪的梦，梦到我去Oracle上班，刚座下来发现Edison座我对面，我好奇的问：您是CHO吗，对方说：是的。搞得我一阵狂喜，我说：您怎么会在这里上班呢？

都怪我，Oracle 3面失败后老做奇怪的梦。。。

ifu

yyzjp 发表于 2011-1-24 10:35
昨天晚上做了个奇怪的梦，梦到我去Oracle上班，刚座下来发现Edison座我对面，我好奇的问：您是CHO吗，对方说 ...

找个心理医生好好看看

yyzjp

ifu 发表于 2011-1-24 11:12
找个心理医生好好看看

哈哈，别这样啊。

Thomson Reuters 的3面成功了，改天就高高兴兴去上班咯。。。所谓东方不亮西方亮。。。

我说的是事实

看到标题就猜是mlaa，进来一看果不其然

Edison

白皮书出来了，从对比图片看，不会像 MLAA 那样对几何体内部的纹理造成干扰：

Edison

2.X MB 的 PNG 格式图片（截取自 Crysis 2）：

http://graphics.cs.williams.edu/papers/SRAAI3D11/crysis2.png

图片说明：

This scene contains many sources of geometric aliasing with low shading frequency, which are particularly well suited for SRAA. For instance, it is able to correctly reconstruct the brick structure along the roof, as well as the bent fence at the far end. The image has been reconstructed using 4 subsamples on an ordered grid to a target resolution of 19201080 pixels in 2.5 ms. Scene courtesy of Crytek from Crysis 2.

Edison

Frostbite 2 游戏引擎（Battlefield 3 采用引擎）截图：

http://graphics.cs.williams.edu/papers/SRAAI3D11/large-detail.jpg

图片说明：
SRAA reconstructs subpixel edges while preserving sharp texture and shading features. The large image was processed by SRAA. The details compare the 1x input, the SRAA output, and the 16x shaded reference. Scene courtesy of DICE from the Frostbite 2 game engine. Zoom the electronic version of this paper to see pixel-scale detail. The reconstruction at the target resolution (19201080) took 2.5 ms.

Edison

作者认为下一个方向是 SRAA 与 MLAA 结合，原因是 MLAA 虽然会造成过分模糊，但是能对着色边缘实现求解，而 SRAA 则不能。

We believe the next step is to combine ideas from SRAA and MLAA. SRAA uses relatively inexpensive geometric information to improve expensive shading results and is able to produce very good edge antialiasing. MLAA’s heuristics often produce overblurring (Figure 2), but it is able to resolve shading edges that SRAA cannot. These include texture, shadow, and specular highlight boundaries. An algorithm that combines heuristic shading weights with accurate geometric weights may be able to achieve higher quality than either alone in practice.