Selection Guide
Public Space Crowd Analysis: Tackling Occlusion and Missed Detections
Dense crowd flow analysis in European public safety scenarios, powered by NVIDIA Jetson Orin NX, focusing on occlusion and missed-detection challenges.
Intelligent Video Analytics
Object Detection · OCR · Anomaly Detection — Edge AI
Local AI for every camera — real-time inference, privacy-safe. From 1-stream edge boxes to 20+ channel hubs, choose compute & deploy in minutes.
Scenario Overview
Edge AI for every camera — object detection, OCR, anomaly discovery. Local inference, real-time, privacy-safe. 1–20+ streams, offline-capable. Deployable across industrial, retail, security & healthcare.
Local Real-Time Inference
- NPU/GPU acceleration, ms response
- Video stays on-device, privacy compliant
- Runs offline without network
Multi-Stream Scalability
- 1–20+ channels at 1080p per device
- Tiered compute: 26 TOPS to 275 TOPS
- Horizontal scaling, no bottleneck
Bandwidth-Efficient & Privacy-Safe
- Metadata only — tens of KB per event
- On-device redaction keeps data local
- Dual RTSP / WebRTC, flexible output
Application Scenarios
Real-time detection, counting & tracking
YOLO-series models detect people, vehicles & objects, with tracking for entry/exit counts, dwell time & heatmaps. The foundational capability behind most vision applications.
Core Advantages
- 80+ common classes out of the box, fine-tune for custom categories
- 1–20+ channels at 1080p real-time per device
- Tracking, counting, and heatmap post-processing with one click
Person / Vehicle / Object Detection
80+ COCO classes, fine-tunable
Entry/Exit Counting & Tracking
Bi-dir count, ID tracking, BI-ready
Zone Heatmap / Line Crossing
Zone heatmaps & line-cross alerts
On-device OCR & QR — image to data
Edge OCR for chars, digits, serials & QR codes + rules engine. Ideal for industrial batch traceability, license plate recognition & production line serial validation.
Core Advantages
- Chinese, English, digits, and QR codes — all-in-one recognition
- Edge inference keeps sensitive text on-device
- Layer rules engine for format validation and database entry
Industrial Batch Codes
Line code OCR, auto-verified
License Plate / Container ID
Gate & container ID, fast trigger
QR / Barcode
QR auto-register, barcode verify
Defects, violations & intrusion detection
Anomaly detection, classification & segmentation models find 'missing' or 'should-not-be-there' targets — surface defects, missing PPE, intrusions, falls. Deploy with few-shot fine-tuning.
Core Advantages
- Supervised classification + unsupervised anomaly detection — dual approach
- Fine-tune custom models with just 50 images on SenseCraft
- Event-triggered local alerts, only tens of KB metadata transmitted
Surface Defect Inspection
Scratch, missing part & misassembly
Safety Violation Alerts
Missing PPE & danger zone alerts
Posture Anomaly
Fall & loiter via 17/33 keypoints
Deployment & Selection
Where to place compute: 3 deployment topologies
A video analytics solution combines four types of components — cameras, AI compute nodes, network, and cloud. Depending on where compute resides, deployments fall into three topologies.
Core Advantages
- All-in-One AI Camera (reCamera / R2130): 1 stream, simplest, plug-and-play
- Detached Edge Box (J3011 / J4012): 2–6 channels, cost-effective workhorse
- Multi-Stream Video Hub (J5012 / AGX Orin): 9+ channels, centralized RTSP processing
| Product Image | Product | Core Positioning | AI Performance (INT8) | Channels (YOLOv8m) | Cost |
|---|---|---|---|---|---|
 | Vision V2 | OEM Custom Board | 0.04 TOPS | — (tiny models only) | ~$30 |
 | reCamera | All-in-One AI Camera | 1 TOPS | 1 ch | ~$80 |
 | R2130 | RPi 5 + Hailo-8 | 26 TOPS | 2–3 ch | ~$200 |
 | J3011 | Jetson Orin Nano 8G | 40 TOPS | ~4 ch | $599 |
 | J4012 | Jetson Orin NX 16G | 100 TOPS | ~7 ch | $949 |
 | J5012 | AGX Orin 64G | 275 TOPS | ~15 ch | $3149 |
Right-sizing: choose hardware by channel count
To determine how many 1080p30 streams you can run, look at TOPS and model size. The table below uses **YOLOv8m** (industrial-grade accuracy) as the common baseline, summarizing AI performance, channel count, and cost across six products.
Core Advantages
- Channel benchmark: YOLOv8m, 1080p @ 15 FPS INT8 (640×640); lighter YOLOv8n/s multiplies channels 2–3×, 30+ FPS halves estimate
- Bottleneck: video decode, not inference — larger models amplify; real-world figures run slightly below pure-inference estimates
- How to choose: 1 ch — reCamera; 2–6 ch — J3011/J4012; 9+ ch or GMSL — J5012
| Product Image | Product | Ethernet | USB | MIPI CSI | GMSL | RTSP Decode (1080p30, H.265) |
|---|---|---|---|---|---|---|
| reCamera | ×1 (100M) | ×1 | — | — | 1 stream (on-device) |
| R2130 | ×1 (GbE) | ×2 | ×2 | — | ~4 streams (RPi 5 VPU) |
| J3011 | ×1 (GbE) | ×4 | ×2 | — | 11 streams |
| J4012 | ×1 (GbE) | ×4 | ×2 | — | 18 streams |
| J5012 | ×4 GbE + ×1 10GbE | ×4 | — | ×8 | 44 streams |
Which device works with my camera?
The table below summarizes video input interfaces and hardware decoding limits for each device.
Core Advantages
- IP Cameras: RTSP via Ethernet, channels limited by HW decode ceiling (NVIDIA data)
- Local Cameras: MIPI CSI (J3011 / J4012) or GMSL (J5012 only, ×8 daughter)
- Output + Industrial Bus: RTSP / WebRTC output; J4012 / J5012 include CAN + RS-485
| Product Image | Product | Cooling Method | Suitable Environment | Operating Temperature | Power Supply |
|---|---|---|---|---|---|
 | reComputer AI Industrial R2135 | Active (Fan) | Standard Indoor | -20~65°C | DC 12-19V |
 | reComputer Industrial R2235 | Passive (Fanless) | Dusty/Workshop | -20~50°C | DC 9-36V Wide |
 | reComputer Super J3011 | Active (Fan) Super | Standard Indoor | -20~60°C | DC 12-19V |
 | reComputer Industrial J3011 | Passive (Fanless) Industrial | Dusty/Workshop | -20~60°C | DC 12V~24V |
 | reComputer Super J4012 | Active (Fan) Super | Standard Indoor | -20~60°C | DC 12-19V |
 | reComputer Industrial J4012 | Passive (Fanless) Industrial | Dusty/Workshop | -20~60°C | DC 12V~24V |
| reComputer Robotics J5012 | Active (Fan) | Standard Indoor / GMSL | -10~60°C | DC 19~48V |
 | reServer Industrial J501 | Passive (Fanless) Industrial | Dusty/Workshop | -20~60°C | DC 12V~36V |
Passive vs Active: choose by dust level
Dust level determines cooling: fan (active) for lower cost & sustained full-load; fanless sealed enclosure (passive) for dust, wide temp & voltage — built for harsh environments.
Core Advantages
- High dust (mining, mills, outdoor) → passive Industrial/Fanless, sealed fanless
- Indoor (office, retail, lab) → active cooling (Super/AI), fan-cooled better value
- Naming: J-series 'Industrial' = passive; R2000 'AI' = fan, else fanless
Success Cases
Contact Us
We Are Glad to Be Your Hardware Partner !