# Showing a captured video sequence with annotations

**URL:** <https://discourse.julialang.org/t/showing-a-captured-video-sequence-with-annotations/2625>\
**Category:** Visualization\
**Tags:** question, images, glvisualize, plotting\
**Created:** [March 12, 2017, 5:36pm UTC](https://discourse.julialang.org/t/showing-a-captured-video-sequence-with-annotations/2625 "2017-03-12T17:36:30Z")\
**Posts on this page:** 2\
**Page:** 2

<div class="post-metadata">

**Author:** ![sdanisch](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/sdanisch/32/1406_2.png) [@sdanisch](https://discourse.julialang.org/u/sdanisch)\
**Post date:** [March 15, 2017, 8:36am UTC](https://discourse.julialang.org/t/showing-a-captured-video-sequence-with-annotations/2625/21 "2017-03-15T08:36:17Z")

</div>

That seems roughly like a simplified form of what I have in GLAbstraction 😉  
What was the main thing missing for you? Maybe documentation? 😃 E.g. automatic conversion is implemented by gl\_convert. I haven’t fully implemented checking what uniforms/attributes a shader needs, since most of my shader use templating and insert the uniforms/attributes automatically.  
I must admit, that the API has some horrible inconveniences, simply because I don’t care that much about it and it works fine in GLVisualize. A big refactor is on the way, hopefully removing these problems and making things faster and consistent!

---

<div class="post-metadata">

**Author:** ![TsurHerman](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/tsurherman/32/1234_2.png) [@TsurHerman](https://discourse.julialang.org/u/TsurHerman)\
**Post date:** [March 15, 2017, 9:44am UTC](https://discourse.julialang.org/t/showing-a-captured-video-sequence-with-annotations/2625/22 "2017-03-15T09:44:53Z")

</div>

Yes in the beginning I thought GLAbstraction is all I needed, but it turned out there is no simple alternative  
to writing shaders by hand in order to achieve various non-orthodox functionality.

I find that it is really easy to work that way. Instead of abstracting away shaders , I am only abstracting away  
the cumbersome and highly annoying openGL C api.

I think there is value in layers of abstraction … openGL C api → ModernGL + GLFW → Shaders → GLabstraction(Templated shaders for primitives and lighting) → GLvisualize → GLplots

this way anyone can choose the level in which they would like to tinker with things.

openCL/CUDA can benefit from this workflow as well. until the grand vision of using LLVM technology directly to write  
kernels from julia code. it would be productive to have a one-liner to compile a kernel into a function and a one-liner to invoke that kernel with automatic conversions from Julia types

[Previous page](https://discourse.julialang.org/t/showing-a-captured-video-sequence-with-annotations/2625.md?page=1)