Revision as of 10:09, 27 June 2024 by Registered User
Applicable for | STM32MP13x lines, STM32MP15x lines, STM32MP25x lines |
Only for STM32MP2 1
|
||||
---|---|---|---|---|
NPU/GPU | CPU | CPU | EdgeTPU | |
Image classification |
||||
Object detection |
||||
Face recognition |
||||
Pose estimation |
||||
Semantic segmentation |
Important |
1 NPU/GPU acceleration based on OpenVX backend using network binary graph (NBG) is available only for STM32MP2x boards |
1. Performance Comparison[edit source]
Model | STM32MP25 NPU |
STM32MP15x CPU | |
---|---|---|---|
Image classification |
MobilenetV1_0.5 128x128 |
515 FPS |
22 FPS |
MobilenetV2_1.0 224x224 |
72 FPS |
2 FPS | |
Object detection |
SSD_MobilenetV1_1.0 300x300 |
40 FPS |
2 FPS |
SSD_MobilenetV2_FPNLite 224x224 | 36 FPS | 0.8 FPS | |
Face recognition |
STModel_FaceReco |
15 FPS |
5 FPS |
Pose estimation |
YoloV8n 256x256 |
59 FPS |
1.3 FPS |
Semantic segmentation |
DeepLabV3 257x257 |
17 FPS |
1 FPS |
Neural processing unit - NEW
Graphics processing units
Central processing unit
Frames per second