Back to the App List

PoseNet for Multi-Person Pose Detection

Lightweight and quantized PoseNet model for real-time multi-person pose estimation on edge hardware.

$0.00
More Information
PoseNet for Multi-Person Pose Detection
app details
Model Type

Pose Estimation

Framework

TensorFlow, TensorFlow Lite, LiteRT, OpenVINO

Algorithm

PoseNet

Technical Details

Compatible with real-time inference pipelines and optimized for CPU/NPU acceleration on SECO platforms to estimate human body keypoints from a single RGB image. It supports multiple persons per frame and outputs skeleton data for each. Inspired by "PersonLab" (Papandreou et al.) and optimized for edge inference.

Use Cases
Industrial Safety and Surveillance
Supported IoT Devices
AI optimized hardware Fanless embedded PCs Modules
supported chipsets
Intel® Core™ i3 Qualcomm® Dragonwing QCS6490 Qualcomm® Dragonwing QCS5430 NXP i.MX 8M Plus NXP i.MX 95
Performance Benchmark
Hardware
Latency (ms)
Memory Usage
Processing unit
SECO SOM-SMARC-QCS6490
208.57 ms
78.91 MB
CPU
SECO SOM-SMARC-iMX8Plus
1646 ms
78.85 MB
CPU
SECO SOM-SMARC-MX95
754.16 ms
76.46 MB
CPU
SECO Titan 300 TGL-UP3 AI
78.02 ms
78.29 MB
CPU
SECO SOM-SMARC-iMX8Plus
1204 ms
77 MB
NPU

Related models