VirtualDub MSU Smart Deblocking Filter
MSU Graphics & Media Lab (Video Group)
Take a look at this article on the new site! Follow the link
https://videoprocessing.ai/video_filters/deblocking-smartdeblocking.html
Algorithm: Dr. Dmitriy Vatolin,
Anton Obuhov, Sergey Grishin
Implementation: Anton Obuhov
About
Filter is intended for recovering quality of video ripped from DVD (for example, when it contains 4 hours of video data), VideoCD, shot on cameraphone or after decompressing by DivX, XviD, H.26x (H.261, H.263, H.264). It performs the accurate blockiness artifact removal, which appears as the grid of squares laid over the video.
Filter automatically determines the blockiness' strength on the frame and in the concrete part of it, the processed video contains maximum details from the original video. The blockiness criterions guarantee the accurate work of every filter component in both hard and soft blockiness cases. Thus, filter will process marginally when on scenes with slow motion, while it will smooth very hard in the case of fast motion in the same film.
The main advantages of the filter are:
- The processing raises the PSNR quality metric of processed video in comparison with the original video
- Real-time processing on 2*CIF video
- Improved hard blockiness processing (video shot by cameraphones or web cameras)
- The unaligned blockiness processing option
- Simple and convenient user interface with the on-the-fly filter processing results in the Preview window
Settings
Filter's interface
Filter's interface allows to adjust the main parameters of processing.
Tip: The "Preview" button allows to
instantly observe the results of filter settings modification.
Tip: You can speed up the processing by
unchecking some options.
Use block preprocessor
Enables block preprocessing. You can see the example of degraded movie with a
lot of blockiness, deblocked with preprocessing and without it:
Decompressed frame
Without block preprocessing
With block preprocessing
Preprocess unaligned
In case of slow motion unaligned blockiness may appear. Such blockiness is the
problem for the most of deblocking techniques. If the option "Preprocess
unaligned" is checked, the filter searches areas with unaligned blockiness and
processes them. There are some examples below:
Decompressed frame
Without unaligned
blockiness processing
With unaligned
blockiness processing
Preprocess chroma
Enables preprocessing of color planes.
Decompressed frame
Without chroma preprocessing
With chroma preprocessing
Process chroma
Enables processing of color planes.
Decompressed frame
Without chroma processing
With chroma processing
Using AviSynth
Filter supports Job Control, and it can be used from AviSynth.
Example:
LoadVirtualDubPlugin("...\MSU_Smart_Deblocking.vdf","MSU_Smart_Deblocking", 0)
clip=AVISource("source.avi", false, "RGB24")
clip.ConvertToRGB32.MSU_Smart_Deblocking(1, 1, 1, 1)
The values passed to the filter in line 3:
1) Enable block preprocessor (0,1)
2) Preprocess unaligned (0,1)
3) Preprocess chroma (0,1)
4) Process chroma (0,1)
Examples
Let's take a look on how filter processes the test sequence "Battle"
(XviD, 1-pass, CBR, 100kbs):
Decompressed frame
MSU Deblocking
MSU Smart Deblocking
The next sample is an example of processing the sequence "flower"
(XviD, 1-pass, CBR, 100kbs):
Decompressed frame
MSU Deblocking
MSU Smart Deblocking
Download
Change log
MSU Smart Deblocking 0.8:
Improved block preprocessing algorithm
Improved main deblocking algorithm
The processing improves the PSNR quality metric of uncompressed and the processed videos
Speed and memory optimization, in comparison with the version 0.5:
? without preprocessing - up to 2.7 times
? with preprocessing - up to 2.4 times
Real-time processing of 2*CIF video
Simple user interface
Removed the "directional filtration" option
Fixed problem with yellow hue of the processed video
MSU Smart Deblocking 0.5:
Improved main deblocking algorithm
Improved color processing algorithm
Improved blockiness preprocessor
Improved unaligned blockiness preprocessor
Added directional filtering control (on/off)
Added save/load directional filtering parameters in job control
Added Edge Sensitivity control
Fixed blockiness creation bug in flat background areas
Minimal speed optimization
MSU Smart Deblocking 0.1:
First public release
See also
Another resources
Video resources:
Projects on 3D and stereo video processing and analysis
- MSU S3D-video analysis reports
- MSU 3D Devices Testing
- 3D Displays Video Generation
- 3D Displays Video Capturing
- Stereo Video Depth Map Generation
- SAVAM Saliensy-Aware Video Compression & Dataset
- Video Matting Benchmark
- Video Inpainting Benchmark
MSU Video Quality Measurement tools
Programs with different objective and subjective video quality metrics implementation
- MSU Video Quality Measurement Tool - objective metrics for codecs and filters comparison
- MSU Human Perceptual Quality Metric - several metrics for exact visual tests
Objective and subjective quality evaluation
tests for video and image codecs
- MSU Video Codecs Comparison 2025
- MSU Video Codecs Comparison 2023-2024
- MSU Video Codecs Comparison 2022
- MSU Video Codecs Comparison 2021
- MSU Video Codecs Comparison 2020
- MSU Cloud Benchmark 2020
- Cloud Encoding Services Comparison 2019
- HEVC/AV1 Codec Comparison 2019
- HEVC/AV1 Codec Comparison 2018
- HEVC/AV1 Codec Comparison 2017
- HEVC Codec Comparison 2016
- HEVC Codec Comparison 2015
- 9-th MPEG4-AVC/H.264 Comparison
- 8-th MPEG4-AVC/H.264 Comparison
- 7-th MPEG4-AVC/H.264 Comparison
- 6-th MPEG4-AVC/H.264 Comparison
Here are available VirtualDub and AviSynth filters. For a given type of digital video filtration we typically develop a family of different algorithms and implementations. Generally there are also versions optimized for PC and hardware implementations (ASIC/FPGA/DSP). These optimized versions can be licensed to companies. Please contact us for details via video(at)graphics.cs.msu_ru.
- MSU Cartoon Restore
- MSU Noise Estimation
- MSU Frame Rate Conversion
- MSU Image Restoration
- MSU Denoising
- MSU Old Cinema
- MSU Deblocking
- MSU Smart Brightness and Contrast
- MSU Smart Sharpen
- MSU Noise generation
- MSU Noise estimation
- MSU Motion Estimation Information
- MSU Subtitles removal
- MSU Logo removal
- MSU Deflicker
- MSU Field Shift Fixer AviSynth plug-in
- MSU StegoVideo
- MSU Cartoonizer
- MSU SmartDeblocking
- MSU Color Enhancement
- MSU Old Color Restoration
- MSU TV Commercial Detector
- MSU filters FAQ
- MSU filters statistics
We are working with Intel, Samsung, RealNetworks and other companies on adapting our filters other video processing algorithms for specific video streams, applications and hardware like TV-sets, graphics cards, etc. Some of such projects are non-exclusive. Also we have internal researches. Please let us know via video(at)graphics.cs.msu_ru if you are interested in acquiring a license for such filters or making a custom R&D project on video processing, compression, computer vision.
- 3D Displays Video Generation
- 3D Displays Video Capturing
- Stereo Video Depth Map Generation
- Automatic Objects Segmentation
- Semiautomatic Objects Segmentation
- New Frame Rate Conversion
- New Deinterlacer
- MSU-Samsung Deinterlacing Project
- Digital TV Signal Enhancement
- Old Film Recovery
- Tuner TV Restore
- Panorama
- Video2Photo
- SuperResolution
- SuperPrecision
- MSU-Samsung image and video resampling
- MSU-Samsung Frame Rate Conversion
- Motion Phase filter
- Deshaker (video stabilization)
- Film Grain/Degrain filter
- Deblurring filter
- Video Content Search
Different research and development
projects on video codecs
- MSU Lossless Video Codec (Top!)
- MSU Screen Capture Lossless Codec (Top!)
- MSU MPEG-2 Video Codec
- x264 Codec Improvement
Other information
- Crazy gallery (filters screams :)
- License for commercial usage of MSU VideoGroup Public Software (please be careful: some soft like metrics has another license!)
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