MSU-Samsung Deinterlacing Project
- Method ideas, project: Dr. Dmitriy Vatolin
- Implementation, ideas: Denis Kubasov, Sergey Putilin, Maxim Makhinya
The filter is designed to suppress comb effect, also known as interlacing, in motion pictures. As opposed to other available filters, MSU Deinterlacing uses a balanced combination of 8 filters to ensure better picture quality.
Examples
We shall compare MSU Deinterlacing Filter to Smart Deinterlace (one of the best available filters written by Donald Graft) using 3 test sequences.
The first example is frame No. 9 from sequence ‘Helicopter’.
A frame from the sequence |
A fragment of the frame |
The fragment processed by MSU Deinterlacing |
The fragment processed by Smart Deinterlace |
It’s easy to see that image feature outlined in red looks much better on the left picture. Smart Deinterlace made it blocky, while MSU Deinterlacing did better job at image reconstruction.
The next example:
A frame from the sequence
The frame processed by MSU Deinterlacing
The frame processed by Smart Deinterlace
Consider fragment of the frame:
The fragment processed by MSU Deinterlacing |
The fragment processed by Smart Deinterlace |
The next example is fragment of frame No. 92 from Schumacher sequence.
A fragment of the frame
The fragment processed by MSU Deinterlacing
The fragment processed by Smart Deinterlace
Frames from sequences tennis & mobl:
AlparySoft, Decomb and MSU SMC deinterlacer.
AlparySoft, Decomb, MSU
Download
This filter was developed from October 2002 to May 2003 by group and from May 2003 in collaboration with SAIT (Samsung Advanced Institute of Technology) so currently filter is unavailable due to contract limitations.
- Deinterlacing filters comparison on tennis (DivX 4, 1.4Mb)
- Deinterlacing filters comparison on mobl (DivX 4, 1.8Mb)
E-mail: video@graphics.cs.msu.ru
Please read MSU filters FAQ before mailing.
New Deinterlacing project: MSU Deinterlacer
-
MSU Benchmark Collection
- Super-Resolution for Video Compression Benchmark
- Video Colorization Benchmark
- Defenses for Image Quality Metrics Benchmark
- Learning-Based Image Compression Benchmark
- Super-Resolution Quality Metrics Benchmark
- Video Saliency Prediction Benchmark
- Metrics Robustness Benchmark
- Video Upscalers Benchmark
- Video Deblurring Benchmark
- Video Frame Interpolation Benchmark
- HDR Video Reconstruction Benchmark
- No-Reference Video Quality Metrics Benchmark
- Full-Reference Video Quality Metrics Benchmark
- Video Alignment and Retrieval Benchmark
- Mobile Video Codecs Benchmark
- Video Super-Resolution Benchmark
- Shot Boundary Detection Benchmark
- The VideoMatting Project
- Video Completion
- Codecs Comparisons & Optimization
- VQMT
- MSU Datasets Collection
- Metrics Research
- Video Quality Measurement Tool 3D
- Video Filters
- Other Projects