Обработка видео

Конактное лицо: Дмитрий Сергеевич Ватолин (dmitriy@graphics.cs.msu.ru)


VQMT3D Project

VQMT3D (Video Quality Measurement Tool 3D) project was created to improve stereoscopic films. Our aim is to help filmmakers produce high-quality 3D video by finding inexpensive ways of automatically enhancing film quality.
Automatic Binary Segmentation of Moving Objects in Video

Algorithm performs detection and tracking of foreground objects. Each frame of video sequence is subdivided into regions of two types: foreground and background objects. Thus each frame can be represented as combination of two layers. After foreground objects are detected the tracking of them is performed.
Building panorama image from video

Filter is intended for reconstructing panorama of scene by movie. The output of the filter is a picture that contains all the scene. The work of MSU Panorama is based on motion vectors. Therefore its performance exceeds 30 fps, which is very good result.
Converting video to high-resolution photo

This filter is intended to create high quality still images from video.
The improvement is achieved by using information from neighbouring frames. The resulting images have also higher resolution than original video.
Hide information in video

MSU StegoVideo allows to hide any file in a video sequence before compression and extract it after decompression with unknown codec.
Moving object phases representation

Filter is intended for representation of moving object phases. It can be useful for sport teaching films building. Sach films allow to accent attention on most important steps of motion.
Old cinema effect for video

The filter is intended to achieve an "Old cinema " effect. There exists an possibility to get the effect of film ageing: color scale adjustment and a set of artifacts to achieve a better realism and to give a "chronicle" view.
Removal of subtitles and logo from video

The filter is intended to remove subtitles, logos and other static opaque objects from video.
Video Frame Rate N-Times Up-Conversion

The algorithm allows to increase video frame rate (fps) by the factor equal to any integer number. The main feature of this method is high robustness with regard to motion complexity
x264 codec improvement

x264 is an open source codec of standard MPEG4 AVC/H.264, developed by programmers from all over the world. Nowadays, x264 is one of the best codecs of its standard, what is verified by different comparisons
Saliency Map Estimation

The goal is to design algorithm for automatic saliency maps construction?for real-life video examples
High quality stereo-to-multiview generation

We have a stereo movie. We want to get a multiview movie. Main problems: dealing with fuzzy edges and filling occlusion areas when extrapolating new views.
Keyframe-based Depth Compression

Keyframe-based Depth Compression with Depth Propagation for Multiview Video Coding
Keyframe-based Depth Compression (Update)

Keyframe-based Depth Compression with Depth Propagation for Multiview Video Coding