Dual YOLOv8n UAV Detection on RK3588S at 42 FPS Using NPU
Real-time YOLOv8n UAV detection on the RK3588S NPU | saturates the sensor at 46 FPS | ~140 MB RAM | fully hardware-accelerated (ISP+RGA+NPU) | on-device LLM scene summaries - alebal123bal/khadas_yolov8n_multithread
Dual YOLOv8n UAV Detection on RK3588S at 42 FPS Using NPU launched on June 14, 2026 with 72 votes, #577 of 22,647 launches that month and more than 95% of that year's launches.
In plain words
This project runs dual YOLOv8n object detection models on an RK3588S processor's neural processing unit, achieving 42 frames per second. It is designed for developers and researchers working with UAV detection systems who need efficient, real-time performance on embedded hardware. The implementation leverages the RK3588S's NPU acceleration to run two detection models simultaneously, making it suitable for applications requiring parallel processing or redundant detection capabilities.
written from the facts on this page · September 2026
From the sources
Single 1280×1280 stream, real-world indoor scene, live on the RK3588S NPU. Real-time YOLOv8n UAV detection at the sensor's 46 FPS ceiling, in ~140 MB of RAM. A high-throughput, low-footprint computer-vision pipeline for the Rockchip RK3588S SoC: it captures live 1080p MIPI frames, runs YOLOv8n across all 3 NPU cores in parallel (lifting throughput from ~31 to 46 FPS — the camera, not the pipeline, is now the limit), and streams the annotated result to HDMI or RTSP. Capture, color-convert/resize and inference run entirely on fixed-function silicon (ISP, RGA, NPU), so the CPU stays free and memory holds flat at ~140 MB per stream — small enough to run on even the cheapest 2 GB RK3588S boards,…from github.com
Does a similar job
all alternatives →
Real-Time YOLO Object Detection in Elixir: Fast, Simple, Extensible2024 · github.com · ▲96A library for object detection and seamless YOLO model integration in Elixir - poeticoding/yolo_elixir
80% faster, 50% less memory, 0% loss of accuracy Llama finetuning2023 · github.com · ▲385Hi HN! I'm just sharing a project I've been working on during the LLM Efficiency Challenge - you can now finetune Llama with QLoRA 5x faster than Huggingface's original implementation on your own local GPU. Some highlights: 1. Manual autograd engine - hand derived backprop steps. 2. QLoRA / LoRA 80% faster, 50% less memory. 3. All kernels written in OpenAI's Triton language. 4. 0% loss in accuracy - no approximation methods - all exact. 5. No change of hardware necessary. Supports NVIDIA GPUs since 2018+. CUDA 7.5+. 6. Flash Attention support via Xformers. 7. Supports 4bit and 16bit…
Clearcam – Add AI object detection to your IP CCTV cameras2025 · github.com · ▲233This runs YOLOv8 + bytetrack with Tinygrad detections (depending on user config) are saved and can be sent to the companion iOS app along with a notification, all video processing is done locally, all footage is encrypted before leaving your computer, and the sending notifications + videos part is optional. This uses tinygrad, so it runs well on my apple silicon macs and should be able to run on a lot of hardware (or will be able to when I remove other deps).
Run Qwen3-Next-80B on 8GB GPU at 1tok/2s throughput2025 · github.com · ▲123Contribute to Mega4alik/ollm development by creating an account on GitHub.
- OOOffline Object Detection and Tracking on a Raspberry Pi2018 · medium.com · ▲106
More ai this month
the category →
Paintings by AI models: programs run through a simulation of oil paint.
AI · 22d ago · stillwet.art
FastRouter.ai▲364Route requests to the right LLM for cost, latency & quality
AI · 23d ago · fastrouter.ai
Source: Show HN post ↗
Launched alongside, June 2026
the whole month →
Fundraisly▲1,537AI fundraising agent that finds investors and books meetings
AI · Jun 2026 · fundraisly.com
Homebrew 6.0.0▲1,481Today, I’m proud to announce Homebrew 6.0.0. The most significant changes since 5.1.0 are a new tap trust security mechanism, the new faster, smaller, default internal Homebrew JSON API, sandboxing on Linux, better defaults informed by our user survey, many brew bundle improvements, improved performance and initial support for macOS 27 (Golden Gate). Happy to discuss any questions here!
Dev tools · Jun 2026 · brew.sh
Hi HN! I was super frustrated at apps like tiktok and wechat preventing me from saving or downloading the animated gifs, so I made a little tool to turn a screen recording into a slack-sized emoji. Hope you get some value from this!
Life & fun · Jun 2026 · vorpus.github.io
Search 18 years of Hacker News and chart how any topic, tool, or person trended. Overlay terms to compare their rise and fall across 45M posts and comments.
Life & fun · Jun 2026 · hackernewstrends.com