MSU Analysing and Tuning of Video Codecs
Codec Analysis
One of the main tasks during codec’s development and tuning is evaluation of compressed video quality and measurements’ data management. We test and develop video codecs of different types and standards for more than 5 years. Some of our publicly available codecs comparison reports are located and can be freely downloaded from videocodecs comparison webpage. We are making commercial projects on codecs’ testing and tuning and are interested in any possible projects in this area.
x264 Parameters Analysis
x264 is a freeware codec of MPEG-4 AVC/H.264 standard. A number of codec comparisons including our own show that x264 currently is one of the best H.264 codecs.
We made an analysis of x264 parameters. Results of performed analysis clearly show usecases of different codec’s parameters. These results may be of interest to professional users of this codec and codec developers. Also such results are of considerable help for codec’s tuning and improvement.
Comparison report freely available at our website.
Example of x264 parameters analysis
Video Codecs Testing and Tuning Software
There is a common task to analyze a video codec to assess its quality, quality of its functions. For example such a task appears when developing a new codec or tuning an already developed one. Such tasks commonly are solved by manual or semiautomatic testing systems and often these systems have not a sufficient number of functions for complete analysis.
MSU Video Quality Metric Tool
Main task of MSU Video Quality Metric Tool (VQMT) is to compare video sequences using objective quality metrics. Main features of proposed system are:
- Support of most popular file formats: AVI, YUV, MOV, MP4,…
- Support of many popular video quality metrics: PSNR, MSE, SSIM, VQM, …
- Graphic user and command-line interfaces
- Plug-ins support
There are two versions of VQMT: free version for personal usage (reduced functionality) and commertional version for companies (all features are included). See VQMT page for more information.
Video Codecs Scoring System
Main tasks of Video Codecs Scoring System (ViCoS) are analysis of video codecs performance based on objective video quality metrics.
Main methods of analysis are:
- Compare codecs based on Rate-Distortion curves for several sequences
- Natural sequences modiffications usage
- Synthetic sequences usage
- Codecs’s presets automatic analysis and selection
Main features of the ViCoS System are the following:
- Easy massive codecs launches using CLI
- Possibility of massive paralles launches using client-server archtecture
- Support of relational data bases
- MATLAB visualization toolbox
More information can be found at ViCoS System page.
Contacts
-
MSU Benchmark Collection
- Video Colorization Benchmark
- Super-Resolution for Video Compression 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