Hacker News
Productive Procrastination
Productive Procrastination
Productive Procrastination
I gave every train in New York an instrument
Understanding the Kalman filter with a simple radar example
Understanding the Kalman filter with a simple radar example
Understanding the Kalman filter with a simple radar example
Show HN: I pipe free sports streams into Jellyfin – no ads, just HLS
Show HN: I pipe free sports streams into Jellyfin – no ads, just HLS
Understanding the Kalman filter with a simple radar example
How Complex is my Code?
X / Twitter
Software Architecture Guide. By Martin Fowler.
https://t.co/AyV3VHiAt0
It's 1:30 am and I've nothing better to do, so here we go:
* stereo vs. mono: yes, if you have stereo, use stereo. If you're stuck with mono and have some compute to spare, use an ML depth model to get approximate depth. Accuracy at this stage is overrated. I'll all get bundle-adjusted anyway.
* feature tracking: search vs. KLT. Both have upsides and downsides. The best systems use a hybrid. Search is often too expensive and depends on detection, which is often brittle -- but can save your butt in many edge cases. KLT is fast & robust but bare KLT is not very accurate and also drifts over time. I hope to be doing one or more posts on this very topic hopefully soon.
* pose estimation: directionally correct but there's a world of best practices to make this fast & robust. People have written entire PhD theses about this. Topic for another post.
* KF-based map expansion: yes that's best practice. But KF-selection based on "every few meters" is instant game over. Lots of cases and edge cases that need to go into a suitable heuristic.
* CUDA kernels for stereo matching: serious overkill. Matching few hundred features on CPU takes at most a couple milliseconds if implemented right.
* Local BA: 12 KFs is kinda arbitrary. Might work well for KITTY but not generalize.
* Eval on KITTY: that's easy-tier: camera always upright. No pure rotations. Very controlled motion. Very large field of view. Drone datasets are where the rubber meets the road.
* Performance: 9 FPS on RTX 3050. NGL, that is brutally, ludicrously slow. Us old-schoolers did realtime visual SLAM 20 years ago on ~ 1/1000th of the compute budget.
As a designer, I truly dislike moodboards, specifically moodboards that are just full of end work from the same industry (ie. moodboards of UIs from other apps)
Inspiring by showing other's work is a depressing practice. At least show something wacky, use movies, use vibes..
sam3.cpp - Meta's SAM 3 in pure C++ with @ggerganov's ggml
- Supports SAM 3.1, 3, 2.1, 2 and EdgeTAM
- FP16, 4-bit quant (EdgeTAM in 15 MB)
- Apple Metal GPU, CUDA, CPU
- Text-prompted: "peach" → every peach
- Single-file C++14
Performance-wise:
- 100ms object detection, segmentation
- Video object segmentation @ 20FPS on M4 Pro with EdgeTAM
https://t.co/XHC7ipyQtI
recommended reading. https://t.co/qq3Zz5TA7y
https://t.co/RZlp39Wkrf
https://t.co/5nQwqqW4EC