yolo-detection-2026-openvino

Name: yolo-detection-2026-openvino
Author: SharpAI

2.9kstars460forks

How to install yolo-detection-2026-openvino?

sharpai/deepcamera/yolo-detection-2026-openvino

$npx -y skills add SharpAI/DeepCamera --skill yolo-detection-2026-openvino

Installs into the current project.

Summary

OpenVINO — real-time object detection via Docker (NCS2, Intel GPU, CPU)

Files · 1

View on GitHub

SKILL.md

1	# OpenVINO Object Detection
2
3	Real-time object detection using Intel OpenVINO runtime. Runs inside Docker for cross-platform support. Supports Intel NCS2 USB stick, Intel integrated GPU, Intel Arc discrete GPU, and any x86_64 CPU.
4
5	## Requirements
6
7	- Docker Desktop 4.35+ (all platforms)
8	- Optional hardware: Intel NCS2 USB, Intel iGPU, Intel Arc GPU
9	- Falls back to CPU if no accelerator present
10
11	## How It Works
12
13	```
14	┌─────────────────────────────────────────────────────┐
15	│ Host (Aegis-AI) │
16	│ frame.jpg → /tmp/aegis_detection/ │
17	│ stdin ──→ ┌──────────────────────────────┐ │
18	│ │ Docker Container │ │
19	│ │ detect.py │ │
20	│ │ ├─ loads OpenVINO IR model │ │
21	│ │ ├─ reads frame from volume │ │
22	│ │ └─ runs inference on device │ │
23	│ stdout ←── │ → JSONL detections │ │
24	│ └──────────────────────────────┘ │
25	│ USB ──→ /dev/bus/usb (NCS2) │
26	│ DRI ──→ /dev/dri (Intel GPU) │
27	└─────────────────────────────────────────────────────┘
28	```
29
30	1. Aegis writes camera frame JPEG to shared `/tmp/aegis_detection/` volume
31	2. Sends `frame` event via stdin JSONL to Docker container
32	3. `detect.py` reads frame, runs inference via OpenVINO
33	4. Returns `detections` event via stdout JSONL
34	5. Same protocol as `yolo-detection-2026` — Aegis sees no difference
35
36	## Platform Setup
37
38	### Linux
39	```bash
40	# Intel GPU and NCS2 auto-detected via /dev/dri and /dev/bus/usb
41	# Docker uses --device flags for direct device access
42	./deploy.sh
43	```
44
45	### macOS (Docker Desktop 4.35+)
46	```bash
47	# Docker Desktop USB/IP handles NCS2 passthrough
48	# CPU fallback always available
49	./deploy.sh
50	```
51
52	### Windows
53	```powershell
54	# Docker Desktop 4.35+ with USB/IP support
55	# Or WSL2 backend with usbipd-win for NCS2
56	.\deploy.bat
57	```
58
59	## Model
60
61	Ships without a pre-compiled model by default. On first run, `detect.py` will auto-download `yolo26n.pt` and export to OpenVINO IR format. To pre-export:
62
63	```bash
64	# Runs on any platform (unlike Edge TPU compilation)
65	python scripts/compile_model.py --model yolo26n --size 640 --precision FP16
66	```
67
68	## Supported Devices
69
70	\| Device \| Flag \| Precision \| ~Speed \|
71	\|--------\|------\|-----------\|--------\|
72	\| Intel NCS2 \| `MYRIAD` \| FP16 \| ~15ms \|
73	\| Intel iGPU \| `GPU` \| FP16/INT8 \| ~8ms \|
74	\| Intel Arc \| `GPU` \| FP16/INT8 \| ~4ms \|
75	\| Any CPU \| `CPU` \| FP32/INT8 \| ~25ms \|
76	\| Auto \| `AUTO` \| Best \| Auto \|
77
78	## Protocol
79
80	Same JSONL as `yolo-detection-2026`:
81
82	### Skill → Aegis (stdout)
83	```jsonl
84	{"event": "ready", "model": "yolo26n_openvino", "device": "GPU", "format": "openvino_ir", "classes": 80}
85	{"event": "detections", "frame_id": 42, "camera_id": "front_door", "objects": [{"class": "person", "confidence": 0.85, "bbox": [100, 50, 300, 400]}]}
86	{"event": "perf_stats", "total_frames": 50, "timings_ms": {"inference": {"avg": 8.1, "p50": 7.9, "p95": 10.2}}}
87	```
88
89	### Bounding Box Format
90	`[

People who liked this also viewed

6 picks

Type

find-skillsHelps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skSkillsJul 20262.6M27k
grill-meA relentless interview to sharpen a plan or design.SkillsJul 2026633k182k
vercel-react-best-practicesVercel's official collection of agent skillsSkillsJul 2026573k29k
improve-codebase-architectureScan a codebase for deepening opportunities, present them as a visual HTML report, then grill through whichever one you pick.SkillsJul 2026522k182k
setup-matt-pocock-skillsConfigure this repo for the engineering skills — set up its issue tracker, triage label vocabulary, and domain doc layout. Run once before first use of the other engineering skills.SkillsJul 2026454k182k
triageMove issues and external PRs through a state machine of triage roles — categorise, verify, grill if needed, and write agent-ready briefs.SkillsJul 2026424k182k

yolo-detection-2026-openvino

2.9kstars460forks

View on GitHub

How to install yolo-detection-2026-openvino?

sharpai/deepcamera/yolo-detection-2026-openvino

$npx -y skills add SharpAI/DeepCamera --skill yolo-detection-2026-openvino

Installs into the current project.

Summary

OpenVINO — real-time object detection via Docker (NCS2, Intel GPU, CPU)

Files · 1

View on GitHub

SKILL.md

1	# OpenVINO Object Detection
2
3	Real-time object detection using Intel OpenVINO runtime. Runs inside Docker for cross-platform support. Supports Intel NCS2 USB stick, Intel integrated GPU, Intel Arc discrete GPU, and any x86_64 CPU.
4
5	## Requirements
6
7	- Docker Desktop 4.35+ (all platforms)
8	- Optional hardware: Intel NCS2 USB, Intel iGPU, Intel Arc GPU
9	- Falls back to CPU if no accelerator present
10
11	## How It Works
12
13	```
14	┌─────────────────────────────────────────────────────┐
15	│ Host (Aegis-AI) │
16	│ frame.jpg → /tmp/aegis_detection/ │
17	│ stdin ──→ ┌──────────────────────────────┐ │
18	│ │ Docker Container │ │
19	│ │ detect.py │ │
20	│ │ ├─ loads OpenVINO IR model │ │
21	│ │ ├─ reads frame from volume │ │
22	│ │ └─ runs inference on device │ │
23	│ stdout ←── │ → JSONL detections │ │
24	│ └──────────────────────────────┘ │
25	│ USB ──→ /dev/bus/usb (NCS2) │
26	│ DRI ──→ /dev/dri (Intel GPU) │
27	└─────────────────────────────────────────────────────┘
28	```
29
30	1. Aegis writes camera frame JPEG to shared `/tmp/aegis_detection/` volume
31	2. Sends `frame` event via stdin JSONL to Docker container
32	3. `detect.py` reads frame, runs inference via OpenVINO
33	4. Returns `detections` event via stdout JSONL
34	5. Same protocol as `yolo-detection-2026` — Aegis sees no difference
35
36	## Platform Setup
37
38	### Linux
39	```bash
40	# Intel GPU and NCS2 auto-detected via /dev/dri and /dev/bus/usb
41	# Docker uses --device flags for direct device access
42	./deploy.sh
43	```
44
45	### macOS (Docker Desktop 4.35+)
46	```bash
47	# Docker Desktop USB/IP handles NCS2 passthrough
48	# CPU fallback always available
49	./deploy.sh
50	```
51
52	### Windows
53	```powershell
54	# Docker Desktop 4.35+ with USB/IP support
55	# Or WSL2 backend with usbipd-win for NCS2
56	.\deploy.bat
57	```
58
59	## Model
60
61	Ships without a pre-compiled model by default. On first run, `detect.py` will auto-download `yolo26n.pt` and export to OpenVINO IR format. To pre-export:
62
63	```bash
64	# Runs on any platform (unlike Edge TPU compilation)
65	python scripts/compile_model.py --model yolo26n --size 640 --precision FP16
66	```
67
68	## Supported Devices
69
70	\| Device \| Flag \| Precision \| ~Speed \|
71	\|--------\|------\|-----------\|--------\|
72	\| Intel NCS2 \| `MYRIAD` \| FP16 \| ~15ms \|
73	\| Intel iGPU \| `GPU` \| FP16/INT8 \| ~8ms \|
74	\| Intel Arc \| `GPU` \| FP16/INT8 \| ~4ms \|
75	\| Any CPU \| `CPU` \| FP32/INT8 \| ~25ms \|
76	\| Auto \| `AUTO` \| Best \| Auto \|
77
78	## Protocol
79
80	Same JSONL as `yolo-detection-2026`:
81
82	### Skill → Aegis (stdout)
83	```jsonl
84	{"event": "ready", "model": "yolo26n_openvino", "device": "GPU", "format": "openvino_ir", "classes": 80}
85	{"event": "detections", "frame_id": 42, "camera_id": "front_door", "objects": [{"class": "person", "confidence": 0.85, "bbox": [100, 50, 300, 400]}]}
86	{"event": "perf_stats", "total_frames": 50, "timings_ms": {"inference": {"avg": 8.1, "p50": 7.9, "p95": 10.2}}}
87	```
88
89	### Bounding Box Format
90	`[

People who liked this also viewed

6 picks

Type