Page content last modified on 2023-09-13.
Kvazaar is an award-winning academic open-source video encoder for the state-of-the-art High Efficiency Video Coding (HEVC/H.265) standard developed since 2012. Kvazaar is being developed in C and optimized in SSE/AVX intrinsics under the BSD-3-Clause license since v2.1. The development is being coordinated by Ultra Video Group and the implementation work is carried out on GitHub.
The main development goals of Kvazaar are:
- Coding efficiency close to HM
- Easy portability to various platforms
- Real- time coding speed
- Optimized computation and memory resources
Kvazaar includes all coding tools of Main, Main 10, and Main Still Picture profiles of HEVC and its modular source code facilitates parallelization on multi and manycore processors as well as algorithm acceleration on hardware. This cross-platform HEVC encoder is targeted at x86, x64, PowerPC, and ARM processors on Windows, Linux, and Mac. Kvazaar is also supported by de-facto standard multimedia frameworks FFmpeg and Libav.
Please cite the following paper (or any from the publication section) for any usage of Kvazaar:
A. Lemmetti, M. Viitanen, A. Mercat, and J. Vanne, “Kvazaar 2.0: fast and efficient open-source HEVC inter encoder,” in Proc. ACM Multimedia Syst. Conf., Istanbul, Turkey, June 2020. [Tuni.fi] [PDF]
Features
Kvazaar is a fully featured HEVC encoder. It supports all main profile tools for intra and inter coding. Kvazaar also supports the following features:
| Quick info | |
|---|---|
| What? | HEVC standard encoder for Windows, Linux and Mac systems |
| Why? | Kvazaar serves as a research platform for new coding tool development and other encoder research activities as well as provides a high-quality HEVC encoder for the public to use |
| How? | Using C (C99) with additional optimizations in SSE/AVX intrinsics |
| Features | |
|---|---|
| License | BSD-3-Clause |
| Intra tools | All Main/Main10 profile |
| Inter tools | All Main/Main10 profile with HEXB, FS, and TZ IME algorithms |
| Input format | YUV420p (and Y4M) |
More details can be found from the GitHub README The development of Kvazaar continues on the side with out VVC encoder uvg266.
Visualizer
Kvazaar comes with a built-in visualizer that can be used to visualize the encoding process. The visualizer is a part of the Kvazaar source code and is available on GitHub.
An example of Kvazaar encoding visualized with the built-in Visualizer:
Publications
P. Sjövall, A. Mercat, and J. Vanne, “FPGA-accelerated HEVC encoder for energy-efficient multi-access edge computing,” in Proc. IEEE Int. Conf. Image Processing, Kuala Lumpur, Malaysia, Oct. 2023. [Tuni.fi] [PDF]
J. Sainio, A. Mercat, and J. Vanne, “RDO candidate selection for maximizing coding efficiency in a practical HEVC encoder,” in Proc. IEEE Int. Conf. Acoustics, Speech, Signal Process., Rhodes Island, Greece, June 2023. [Tuni.fi] [PDF]
A. Mercat, A. Lemmetti, J. Sainio, and J. Vanne, “AVX2-Optimized interpolation filters for HEVC inter encoding,” in Proc. IEEE Int. Symp. Circuits and Syst., Monterey, California, USA, May 2023. [Tuni.fi] [PDF]
A. Mercat, S. Ahovainio, and J. Vanne, “Spatio-temporal parallelization scheme for HEVC encoding on multi-computer systems,” in Proc. IEEE Int. Conf. Image Processing, Bordeaux, France, Oct. 2022. [Tuni.fi] [PDF]
A. Lemmetti, M. Viitanen, A. Mercat, and J. Vanne, “Kvazaar 2.0: fast and efficient open-source HEVC inter encoder,” in Proc. ACM Multimedia Syst. Conf., Istanbul, Turkey, June 2020. [Tuni.fi] [PDF]
A. Mercat, A. Lemmetti, M. Viitanen, and J. Vanne, “Acceleration of Kvazaar HEVC intra encoder with machine learning,” in Proc. IEEE Int. Conf. Image Processing, Taipei, Taiwan, Sept. 2019. [Tuni.fi] [PDF]
J. Sainio, A. Mercat, and J. Vanne, “Hardware deceleration of Kvazaar HEVC encoder,” in Proc. Int. Conf. Embedded Comput. Syst.: Architectures, Model. Simul., Samos, Greece, July 2019. [Tuni.fi] [PDF]
A. Lemmetti, E. Kallio, M. Viitanen, J. Vanne, and T. D. Hämäläinen, “Rate-distortion-complexity optimized coding scheme for Kvazaar HEVC intra encoder,” in Proc. Data Compression Conf., Snowbird, Utah, USA, Mar. 2018. [Tuni.fi] [PDF]
A. Ylä-Outinen, A. Lemmetti, M. Viitanen, J. Vanne, and T. D. Hämäläinen, “Kvazaar: HEVC/H.265 4K30p intra encoder,” in Proc. IEEE Int. Symp. Multimedia, Taichung, Taiwan, Dec. 2017. [Tuni.fi] [PDF]
P. Sjövall, V. Viitamäki, A. Oinonen, J. Vanne, T. D. Hämäläinen, and A. Kulmala “Kvazaar 4K HEVC intra encoder on FPGA accelerated airframe server,” in Proc. IEEE Workshop Signal Process. Syst., Lorient, France, Oct. 2017. [Tuni.fi] [PDF]
M. Viitanen, A. Koivula, A. Lemmetti, A. Ylä-Outinen, J. Vanne, and T. D. Hämäläinen, “Kvazaar: open-source HEVC/H.265 encoder,” in Proc. ACM Int. Conf. Multimedia, Amsterdam, The Netherlands, Oct. 2016. [Tuni.fi] [PDF]
M. Viitanen, A. Koivula, J. Vanne, and T. D. Hämäläinen, “Live demonstration: run-time visualization of Kvazaar HEVC intra encoder,” in Proc. IEEE Int. Symp. Circuits Syst., Montreal, Canada, May 2016. [Tuni.fi] [PDF]
A. Koivula, M. Viitanen, A. Lemmetti, J. Vanne, and T. D. Hämäläinen, “Performance evaluation of Kvazaar HEVC intra encoder on Xeon Phi many-core processor,” in Proc. IEEE Global Conf. Signal Information Process., Orlando, Florida, USA, Dec. 2015. [Tuni.fi] [PDF]
M. Viitanen, A. Koivula, J. Vanne, and T. D. Hämäläinen, “Kvazaar HEVC still image coding on Raspberry Pi 2 for low-cost remote surveillance,” in Proc. IEEE Visual Comm. and Image Proc., Singapore, Dec. 2015. [Tuni.fi] [PDF]
A. Koivula, M. Viitanen, J. Vanne, T. D. Hämäläinen, and L. Fasnacht, “Parallelization of Kvazaar HEVC intra encoder for multi-core processors,” in Proc. IEEE Workshop Signal Process. Syst., Hangzhou, China, Oct. 2015, pp. 1-6. [Tuni.fi] [PDF]
M. Viitanen, A. Koivula, A. Lemmetti, J. Vanne, and T. D. Hämäläinen, “Kvazaar HEVC encoder for efficient intra coding,” in Proc. IEEE Int. Symp. Circuits Syst., Lisbon, Portugal, May 2015, pp. 1662-1665. [Tuni.fi] [PDF]