Have any of you been able to find aftermarket (or official) colored versions of outer key fob trim?
I’m not a huge fan of these overmold style ones.
I’ve been resisting just making one myself…
So technically its not the Kalman part that’s most computationally taxing it’s the o(n3) Hungarian matching I have on top (and now we’re firmly in mumbo jumbo tech speak land :))
I don’t really expect this to be an issue unless if for some reason there are dozens of tracks.
Found the engineer :) The blog post reflects an earlier version of the code. The architecture is now fully decoupled with independent pipelines.
Camera thread (60 fps) —> writes to ring buffer
Ring buffer —> inference thread (variable FPS - in practice 60 fps) —> mailbox
Ring buffer + mailbox...
I would've added it as an option to my build if it was available!
But alas... but then again now I have something I have full control of and can customize however I'd like powered by state of the art hardware + ML models :)
I suspect changing their stack to Android Automotive OS probably had...
Ha - no there's 0 hackery going on here. In fact 0 car-side software changes. This is basically just a fancy dash cam.
What are you thinking? If I had control over the headlight module I'd probably light up detected collision risks.
Everything runs off an Nvidia Orin Nano sitting in the frunk. Using a YOLO model trained on thermal data.
See blog post here (directionally correct; I’ve made a bunch of changes since that aren’t reflected):
https://www.sahouh.com/build-logs/part-5-inferencing-at-60-fps
For those interested...
ha! if you’ve got a few thousand labeled thermal images of kangaroos, I can bake them into the model. the current model’s trained on:
1) People
2) Wild hogs
3) Coyotes
4) Deer
5) Raccoon
6) Bikes
Basically just about everything you might come across in California