Object Detection at 1840 FPS with TorchScript, TensorRT and DeepStream

I see this and I immediately think of &quot;trash sorting&quot; at ultra high speed. If one can combine this with a bunch of accurate (laser precision) air guns, to shoot and move individual pieces of trash you can sort through a truck load of trash in a matter of seconds, perhaps in the air while they are being dumped! compare this approach with how we are currently doing it [0] - Somebody should get Elon Musk on this project right away!<p>[0] - <a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=QbKA9uNgzYQ" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=QbKA9uNgzYQ</a>

Out of curiosity, what are the possible use cases for object detection at &gt;100 fps? I assume it would have to be objects that move very fast, i.e. nothing ordinary that I can think of.<p>[edit] actually stupid question. I assume it&#x27;s more about throughput than fps, i.e. be able to process lots of streams on the same machine, for instance for doing mass analysis of CCTV streams.

How portable are these techniques to other architectures? Could &gt;100 FPS be realistically achieved today using only CPUs or mobile phones?

A weird question, but since there&#x27;s another article on HN right now about programming language energy efficiency <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=24816733" rel="nofollow">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=24816733</a> any idea whether going from 9fps to 1840fps consumes the same power, 200x the power, or somewhere in between?

Good work getting TensorRT running we had a real pain in the butt recently when working with it and just opted to go with ONNXRuntime, their graph optimizer and their TensorRT backend -- may not be as fast as straight TensorRT from comparisons I&#x27;ve seen but it got us to a competitive inference and latency so we&#x27;re happy with it.

Any word on latency? I didn&#x27;t see anything in the article. I guess, since this is a synthetic test just pumping a single image file through repeatedly instead of an actual video stream, then it wouldn&#x27;t realistically be measurable. But if latency is particularly low, this would be a boon for AR systems.

&gt; There is evidence (measured using gil_load) that we were throttled by a fundamental Python limitation with multiple threads fighting over the Global Interpreter Lock (GIL).<p>Can anyone comment on how often this is a problem and if this problem is truly fundamental to Python? Could it be solved in a Python 3.x release?

Name clash again… I thought about <a href="https:&#x2F;&#x2F;deepstream.io&#x2F;" rel="nofollow">https:&#x2F;&#x2F;deepstream.io&#x2F;</a>