| RUSSIAN |
VirtualDub MSU Denoising Filter
(Noise Removal)
MSU Graphics & Media Lab (Video Group)
Algorithm: Dr. Dmitriy Vatolin
Implementation: Sergey Putilin, Viktor Sheludko
New version 1.6.3 available (new denoising algorithm)
The filter is designed for noisy video stream processing, video quality improvement, or video preprocessing (e.g. for preprocessing before compression or deinterlacing).
Settings
The filter suppresses noise in both spatial and temporal domains. Filter effect is specified for each type of denoising in the dialogue box as shown. Checkboxes help to set required combination of domains, thus enabling spatial-only or temporal-only denoising.
"Preview" and "Update Preview" buttons added from version 1.6.2. "Update Preview" button is used to update preview window after changing parameters in parameters window, since it is not done automatically now. I hope we'll implement it in next version :)
VirtrualDub job control added from version 1.6, now filter is usable in AviSynth. Example:
LoadVirtualDubPlugin
("..\VirtualDub\plugins\MSU_denoiser_mmx.vdf","Denoise", 0)
clip=AVISource("D:\work\video\film.avi", false, "RGB24")
clip.ConvertToRGB32.Denoise(50, "20.0")
Firts filter parameter - temporal denoising thresholdSecond filter parameter - spatial denoising intensity
Examples
Let's consider filter operation on Susy sequence first. The sequence is quite noisy – spatial noise is easy to observe, and temporal noise is computed and visualized using LUV metric.
Source frame with noise
LUV metric for frames 13 and 14
It is the plenty of green dots in LUV metric which tells about the presence of temporal noise.
Red spots show where motion takes place.
Let's compare source frame with the frame processed by MSU Denoising Filter:
Source frame with noise
Frame processed by MSU Denoising Filter
(made with 'hard' preset that emphasizes
filter effect)
LUV metric for source and processed frames.
LUV metric
tells that we managed to get rid of the most of spatial noise.
Fragments of the same frame are shown below.
A fragment of source frame
The fragment of the frame processed by MSU Denoising Filter
(made with 'hard' preset that emphasizes filter effect)
And another fragment of the same frame:
A fragment of source frame
The fragment of the frame processed by MSU Denoising Filter
(made with 'hard' preset
that emphasizes filter effect)
Finally, consider MSU Denoising Filter application to Arm sequence from Terminator 2 movie.
A fragment of source frame
The fragment of the frame processed by MSU Denoising Filter
(made with 'hard' preset
that emphasizes filter effect)
Here are results of old (faster) algorithm and new (more quality) algorithm
original image,
parts of image in placeholders are enlarged in the next examples.
part of the frame, processed by old (faster) algorithm
part of the frame, processed by new (more quality) algorithm
part of the frame, processed by old (faster) algorithm
part of the frame, processed by new (more quality) algorithm
part of the frame, processed by old (faster) algorithm
part of the frame, processed by new (more quality) algorithm
Download
- This filter for VirtualDub (79Kb, ZIP). To use, unpack to plugin folder.
- Description and comparison of noise suppression methods used in other available filters.
- Description of
MSU Denoising (free) & MSU Noise Removal filters.
In this report we show how MSU Noise Removal usage can increase compression up to 30% with better (!) visual quality. - Filter license
- This filter for VirtualDub (previous version, 80Kb, ZIP). To use, unpack to plugin folder.
| e-mail: |  |
Other resources
Video resources:
Server size: 8069 files, 1215Mb (Server statistics)
Project updated by
Server Team and
MSU Video Group
Project sponsored by YUVsoft Corp.
Project supported by MSU Graphics & Media Lab