Anti-aliasing may refer to any of a number of techniques to combat the problems of aliasing in a sampled signal such as a digital image or digital audio recording.

Specific topics in anti-aliasing include:

[3] [4] [5]

  • Temporal anti-aliasing (TAA) in CGI, techniques to reduce or remove the effects of temporal aliasing in moving images
    • Deep learning anti-aliasing, a type of spatial and temporal anti-aliasing method relying on dedicated tensor core processors
    • Deep learning super sampling (DLSS), a family of real-time deep learning image enhancement and upscaling technologies developed by Nvidia that are available in a number of video games.

Aliasing & Anti-Aliasing in CGI

edit

The purpose of anti-aliasing is to reduce, if not eliminate, distracting aliasing artifacts in Computer Generated Imagery. Most aliasing artifacts, or jaggies, result from using only one sample point per pixel when computing 2D images. Images produced with single point sampling can achieve the fastest frame rate, at the expense of aliasing artifacts such as “stairsteps” and “narrow faces breakup”. These artifacts are more noticeable in real-time (RT) CGI applications, such as when simulating visual scenes for flight training or playing computer video games. In moving images, stairsteps result in “crawling“. Also, “narrow faces breakup” result in “faces popping in-and-out of scenes”. Most anti-aliasing solutions rely on subpixel processing or postprocessing to reduce stairsteps and crawling. But they have limited success at correcting narrow faces breakup. A new approach, area-based anti-aliasing (ABAA), relies on subpixel area sampling. It is the fastest and produces the best static and moving images with anti-aliasing. Currently, there is no readily available product using ABAA. But simulations have shown that ABAA is the best anti-aliasing solution. Two methods rely on multiple sample points anti-aliasing. It consists of computing multiple images, followed by averaging. The super-sampling anti-aliasing (SSAA) approach has been used in non-real-time applications. In this approach, a 512x512 image is first computed at higher resolution, such as 2048x2048, for example. It is then reduced through averaging or filtering to produce a 512x512 image. It is computation intensive and cannot be used for RT CGI applications. Since there are no time constraints, large images can be computed offline using high-speed general-purpose. With multisample anti-aliasing (MSAA), images are computed for 4 (or 8) subpixel sample points, followed by averaging. It is slow, since the frame rate is reduced by a factor of 4 (or 8). It works well for horizontal and vertical triangle edges. For other edge angles, the gaps between subpixels can cause narrow face breakups. The position of the subpixel sample points for MSAA can be derived from solutions to the “Eight-queens puzzle”. The solutions to the 8-queens puzzle provide good results for near horizontal and vertical edges. But the anti-aliasing effectiveness is not as good for edges with angles in-between. In order to improve the frame rate, several methods rely on single point sampling followed by image post-processing, with mixed results. Among these, there are: Fast approximate anti-aliasing (FXAA), Temporal anti-aliasing (TAA) and Morphological antialiasing (MLAA).

See also

edit

References

edit
  1. ^ "Computer Generated Imagery (CGI), see also www.anti-aliasing.com".
  2. ^ "Conservative Morphological Anti-Aliasing (CMAA)". Intel. Retrieved 2023-03-15.
  3. ^ "New Area-Based Anti-Aliasing for CGI, Michel A. Rohner, Gotham Books Inc 2024-03-15.
  4. ^ “Anti-Aliasing with MSAA vs ABAA”, Michel A. Rohner, Gotham Books Inc 2024-03-15.
  5. ^ “Introduction to Area-Based Anti-Aliasing for CGI”, Michel A. Rohner, Gotham Books Inc 2024-05-15.