- 109comments
- 17comments
- —discuss
- 54comments
- 226comments
- 36comments
- —discuss
- 603comments
- 145comments
- 18comments
- 3comments
- 16comments
- 238comments
- 90comments
- 172comments
- 75comments
- 8comments
- 236comments
- 44comments
- —discuss
- 131comments
- 161comments
- 17comments
- 27comments
- 69comments
- 29comments
- 45comments
- —discuss
- 375comments
- 41comments
Runs so much faster in Chrome than in FF.
which OS, by chance?
Win10
This is amazing. I've been wondering of using such technology in training tennis, so that you get a real-time feedback loop on where you should correct your movement. Will definitely explore it more.
also during workouts
I did something similar for sports (http://www.sportstrace.com), but this looks amazing and (TBH) better! I'm excited to check this out.
Good job. This is impressive. I don't see how parent link is better than yours. Maybe, due to neural nets?
Might be better in that you don't need extra hardware, but i would assume your tracking would be more accurate ?.
I think that would be an amazing comparison between systems to check the accuracy at critical points of inspection!
Link to the demo: https://storage.googleapis.com/tfjs-models/demos/posenet/cam...
Dang it even works on mobile safari (but 2 FPS)
Can it give me a warning signal when I'm slouching?
With some hacking, probably. It'd probably be easiest to have the camera face your side, then just calculate the grades between hips, shoulders, and head. Obviously, you'd have to calibrate it to your posture and it wouldn't be spot on.
It's amazing how much information is conveyed just by a handful of dots in motion. I had a lot of fun a few years ago turning some Michael Jackson moves into an animated constellation: https://codepen.io/sdellis/full/PGLLwx/
Nice. I can almost hear the music.
This is great. I have a huge set of professional creative portraits (>1k finished portraits / >3k unfinished) that are all already indexed for poses with a manual data set. Interested in comparing?