[kekeblom]

Working on making computer vision for robotics easier.

I feel like the real reason we don't have robots doing all of our physical grunt work, is that it is just way too expensive to develop robots that do advanced and custom things. Currently only big players with deep pockets can consider developing software for a robot that automates a task that requires some level of fine grained control or visual feedback.

The computer vision part is challenging as you need deep subject knowledge, you need to implement complicated algorithms and assemble huge datasets. We are working on a way to improve the efficiency of creating datasets, standard algorithm implementations and standardizing interfaces to make things as composable as possible.

It ain't much, but it's a start. https://www.strayrobots.io

[jfim]

That's pretty cool, and best of luck!

Lowering the barrier to entry for many of these object recognition tasks makes so many applications possible. Many years ago (before DL was a thing), I was approached to make a phone app that recognized coins; doing it with classical computer vision, with rotation invariance and potential overlaps between the coins was a rather big task and far too expensive for that customer. Nowadays it's way easier with machine learning, and SDKs such as yours seem to make that even easier.

[high_byte]

this looks awesome.