search

triton-inference-server / dali_backend

The Triton backend that allows running GPU-accelerated data pre-processing pipelines implemented in DALI's python API.
https://docs.nvidia.com/deeplearning/dali/user-guide/docs/index.html
MIT License
120 stars 28 forks source link

Video decode remap #162

Closed szalpal closed 1 year ago

szalpal commented 1 year ago

With https://github.com/NVIDIA/DALI/pull/4613, DALI added a possibility to stream portions of decoded video frames. Please refer to VideoInput operator documentation for more details on this topic.

This feature has also found its way to DALI Backend, especially with #168 and #173 . This example presents, how to use the VideoInput operator together with Triton client sample code. Also, extensive description is provided.

stiepan commented 1 year ago

Nitpick: The name of the PR does not mention it is about adding the usage example. How about adding a two sentence description of why and what is added in the PR?

szalpal commented 1 year ago

Not to reiterate this PR again, I'll refer to these remarks in the followup PR, where I'll be adding this example to the CI.