To apply for this task please complete this mandatory coding challenge together with your application: T872
Applications without matching coding challenge completed will not be considered.
OpenCine currently has capabilites to parse a few RAW file formats such as RAW12 (from apertus cameras), MLV (magic lantern's RAW file format), AdobeDNG etc. However, most editors and video players do not support RAW file formats. Even if they do, they do not work with special data formats like RAW12, since it is not that widely used. We want to use the the capabilities of these advanced editors with file formats that OpenCine can parse already.
For such case so called frame server is used. It is an application which provides video data through some virtual file, e.g. in AVI or MPEG format. Most approaches use FUSE (Filesystem in Userspace) to intercept call to such a file and to provide data to the calling video player/editor.
Additionaly it has to be ensured that memory usage is low enough and performance acceptable. Next step would be a mechanism to provide proxy of the clips for not high-end machines. The focus of this task is to implement OpenCine module, which re-uses OCcore lib to provide video data through libfuse/Dokany as dummy AVI file.
- Create a frame server module using libfuse and attach OCcore as data provider
- Extend OCcore for providing consequent frames, so the video is playable in a video player
- Add user interface to control frame server settings, like frames per second (FPS) or de-Bayering processor
- Add Dokany, to be able to run frame server on Windows and adjust CMake scripts accordingly, for consistent build process without user intervention, like on Linux
- Extend OCcore with LibRaw, to be able to process more RAW formats
You are expected to have working knowledge of (most of) the following:
-- C and C++ and their coding standards (Any IDE)
-- FUSE (file in user space for linux)
-- RAW12, MLV, AVI, BMP, Adobe DNG file formats
-- Qt 5.10 and higher (for creating the User Interface)
-- Dynamic libraries (and their linking - see T872 for appropriate task that tests this)
-- Git based workflow, Github account and Git client
-- Debayering methods and their implementation in OpenCine (T872 tests this)
-- Basic image processing
**Difficulty level:** Medium
**Mentors:** Andrej, Supragya
- Frameserver: https://en.wikipedia.org/wiki/Frameserver
- Open Cine Overview: https://www.apertus.org/opencine
- Wiki: https://wiki.apertus.org/index.php/OpenCine
- FUSE: https://de.wikipedia.org/wiki/Filesystem_in_Userspace
- libfuse: https://github.com/libfuse/libfuse
- Dokany: https://github.com/dokan-dev/dokany
- MLV file format: https://www.magiclantern.fm/forum/index.php?topic=7122.0
- DNG file format: https://helpx.adobe.com/in/photoshop/digital-negative.html
- Frame server prototype (FUSE / AVI): https://github.com/BAndiT1983/OC_FrameServer/tree/dev
To get in touch with any mentor check the [[ https://www.apertus.org/GSoC-2019-Mentor-Contact-List | Mentor Contact List]].