|
Our interests are in areas of format conversion and transcoding
(deinterlacing, frame rate conversion, resampling, super resolution),
video quality enhancement filters (denoising, film grain noise
removing and restoration, deblocking, deblur, old film restoration,
deflicking, sharpen, etc) and other specific filters like creation of
panorama from video, automatic and semiautomatic segmentation, face
detection, etc.
Projects:
- Video Frame Rate N-Times Up-Conversion
- Removal of comb effect from video
- Building panorama image from video
- TV tuner signal recovery
- Converting video to high-resolution photo
- Smart removal of blocking effects from video
- Video Frame Rate Double Up-Conversion
- High-quality blocking reduction
- Old film restoration
- Automatic fixing of shifted fields in video
- Old cinema effect for video
- High-quality increase of video color depth
- High-quality video resolution enhancement
- Creation of cartoons from video
- Hide information in video
- Smart enhance of frame sharpness
- Smart brightness and contrast correction
- Removal of subtitles and logo from video
- Removal of noise from video
- Removing of brightness flicker in video
- Removing of blocking effect from video
In this field our works and main interests are video quality
evaluation, codecs testing, analysis and tuning, lossless compression.
Projects:
- Objective comparison of H.264 video codecs
- Subjective Assessment of Video Quality
- Objective Metrics of Video Quality
- Lossless Video Compression
- Subjective Comparison of Video Codecs
We conduct research into visualization of image-based data obtained
from digital hand-held cameras and laser scanners. Also we have results in
interactive hardware-accelerated graphics and in physically
accurate ray tracing methods (including Monte-Carlo ray tracing).
Projects:
- Image-based rendering
- Physically accurate rendering
|
 |
We conduct research in the fields of face detection and tracking,
facial features detection and 3D shape reconstruction from images.
The applications are: 3D head model reconstruction, model-based video
coding (MPEG-4) and development of natural Human-Computer interface.
Projects:
- Face modeling and recognition
- Natural Human-Computer interface
- Camera Calibration Toolboxes
- Detection of Moving Objects in Video
- Face Tracking in Low Resultion Video
- 3D Object Reconstruction
- "Grow-Cut" Image Segmentation
- Modest AdaBoost
Our interests touch upon fractal image compression, multiresolution
wavelet image compression with aim to high quality and progressive
transmission via networks. Also we are working on specific methods for
texture compression.
Projects:
- Video filtering and measurement
- High quality image and video resampling
- Texture Compression & Synthesis
Our research activity in the domain of image enhancement includes high-quality image resampling and denoising.
Projects:
- Image Resampling Algorithms
- Noise Reduction for Digital Images
In this field our work is developed to visualization of processes with
different nature, such as spatual sound propagation, density etc.
Our main activities are aimed to speech recognition, speaker separation
and indexing. We use neural networks, HMMs, etc.
|