VirtualDub MSU Subtitle Remover
Project, ideas: Dr. Dmitriy Vatolin
Version 1.2 (MSU Subtitle&Logo Remover): Dmitriy Kulikov
Version 2.0b (MSU Subtitle Remover): Dmitriy Kulikov, Sergey Putilin
Version 3.0beta2 (MSU Subtitle Remover): Dmitriy Kulikov, Sergey Putilin, Yuriy Berdnikov
New version of filter for LOGO removal is here MSU Logo Remover
This filter is a plugin for VirtualDub (Free!)
What`s new in 3.0 beta2?
- Fixed bug with "upside-down detection areas"(Thanks for your feedback!)
- Fixed minor bugs with highly compressed video
- New processing mode - single pass, no temporary files. Filter became more user-friendly
What`s new in 3.0 beta?
- New detection algorithm - works up to 2 times faster with less memory usage and better quality
- Improved removing algorithm
Main filter ideas:
- Static subtitles - text doesn`t move for some time
- 2-Pass or single-pass mode - Detection and removing
- Possibility of external mask editing
- Two subtitle removing methods
`Cutted` sequence, 10th frame after subtitle appearing, fast linear camera motion
`House M.D.` sequence, 12th frame after subtitle appearing, scene change
`Friends` sequence, 17th frame after subtitle appearing, camera shaking and slow motion
`Anime3` sequence, 17th frame after subtitle appearing, high camera shaking
Speed comparsion of different versions. Measured on AMD Athlon 2200+
How to speed up filter?
Set `Detection Area` parameter manually (See filter User's Guide for more details) - smaller area means much faster work!
- Unpack this file in VirtualDub "Plugins" folder. File size ~150 KB (in ZIP)
- Example video (6МB)
- Filter User's Guide. Please, read it!
- Filter license
- Prevous stable version 2.2. (size 133 KB)
|For common questions about filter's work and other issues about it please contact us|
|For commercial license of this filter (commercial usage is not free) please contact us|
|Last updated: 09-September-2019|
Project updated by
Server Team and MSU Video Group
Project sponsored by YUVsoft Corp.
Project supported by MSU Graphics & Media Lab