Hugging Face's logo

STMicroelectronics
/
yolov8n

Object Detection
Model card Files
xet
Community
FBAGSTM commited on
Commit
11bffa3
·
verified ·
1 Parent(s): 48fb447

Release AI-ModelZoo-4.0.0

Browse files
Files changed (1) hide show
  1. README.md +37 -15
README.md CHANGED
@@ -65,19 +65,41 @@ https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/
65
 
66
 
67
  ### Reference **NPU** memory footprint based on COCO Person dataset (see Accuracy for details on dataset)
68
- | Model | Dataset | Format | Resolution | Series | Internal RAM | External RAM | Weights Flash | STM32Cube.AI version | STEdgeAI Core version |
69
- |---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-------------|----------|--------------|----------|----------------|----------------|-----------------|------------------------|-------------------------|
70
- | [YOLOv8n per channel](https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/stedgeai_models/object_detection/yolov8n_192_quant_pc_uf_od_coco-person.tflite) | COCO-Person | Int8 | 192x192x3 | STM32N6 | 261 | 0 | 2936.52 | 10.2.0 | 2.2.0 |
71
- | [YOLOv8n per channel](https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/stedgeai_models/object_detection/yolov8n_256_quant_pc_uf_od_coco-person.tflite) | COCO-Person | Int8 | 256x256x3 | STM32N6 | 624 | 0 | 2941.09 | 10.2.0 | 2.2.0 |
72
- | [YOLOv8n per channel](https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/stedgeai_models/object_detection/yolov8n_320_quant_pc_uf_od_coco-person.tflite) | COCO-Person | Int8 | 320x320x3 | STM32N6 | 839.06 | 0 | 2947.02 | 10.2.0 | 2.2.0 |
73
- | [YOLOv8n per channel](https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/stedgeai_models/object_detection/yolov8n_416_quant_pc_uf_od_coco-person.tflite) | COCO-Person | Int8 | 416x416x3 | STM32N6 | 2242.84 | 0 | 2958.34 | 10.2.0 | 2.2.0 |
 
74
  ### Reference **NPU** inference time based on COCO Person dataset (see Accuracy for details on dataset)
75
- | Model | Dataset | Format | Resolution | Board | Execution Engine | Inference time (ms) | Inf / sec | STM32Cube.AI version | STEdgeAI Core version |
76
- |---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-------------|----------|--------------|---------------|--------------------|-----------------------|-------------|------------------------|-------------------------|
77
- | [YOLOv8n per channel](https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/stedgeai_models/object_detection/yolov8n_192_quant_pc_uf_od_coco-person.tflite) | COCO-Person | Int8 | 192x192x3 | STM32N6570-DK | NPU/MCU | 16.88 | 59.24 | 10.2.0 | 2.2.0 |
78
- | [YOLOv8n per channel](https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/stedgeai_models/object_detection/yolov8n_256_quant_pc_uf_od_coco-person.tflite) | COCO-Person | Int8 | 256x256x3 | STM32N6570-DK | NPU/MCU | 24.94 | 40.1 | 10.2.0 | 2.2.0 |
79
- | [YOLOv8n per channel](https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/stedgeai_models/object_detection/yolov8n_320_quant_pc_uf_od_coco-person.tflite) | COCO-Person | Int8 | 320x320x3 | STM32N6570-DK | NPU/MCU | 31.75 | 31.5 | 10.2.0 | 2.2.0 |
80
- | [YOLOv8n per channel](https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/stedgeai_models/object_detection/yolov8n_416_quant_pc_uf_od_coco-person.tflite) | COCO-Person | Int8 | 416x416x3 | STM32N6570-DK | NPU/MCU | 53.11 | 18.83 | 10.2.0 | 2.2.0 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
81
  ### Reference **MPU** inference time based on COCO Person dataset (see Accuracy for details on dataset)
82
  Model | Format | Resolution | Quantization | Board | Execution Engine | Frequency | Inference time (ms) | %NPU | %GPU | %CPU | X-LINUX-AI version | Framework |
83
  |-----------|--------|------------|---------------|-------------------|------------------|-----------|---------------------|-------|-------|------|--------------------|-----------------------|
@@ -94,10 +116,10 @@ Dataset details: [link](https://cocodataset.org/#download) , License [CC BY 4.0]
94
 
95
  | Model | Format | Resolution | AP* |
96
  |-------|--------|------------|----------------|
97
- | [YOLOv8n per channel](https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/stedgeai_models/object_detection/yolov8n_192_quant_pc_uf_od_coco-person.tflite) | Int8 | 192x192x3 | 53.50 % |
98
  | [YOLOv8n per channel](https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/stedgeai_models/object_detection/yolov8n_256_quant_pc_uf_od_coco-person.tflite) | Int8 | 256x256x3 | 58.40 % |
99
- | [YOLOv8n per channel](https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/stedgeai_models/object_detection/yolov8n_320_quant_pc_uf_od_coco-person.tflite) | Int8 | 320x320x3 | 61.80 % |
100
- | [YOLOv8n per channel](https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/stedgeai_models/object_detection/yolov8n_416_quant_pc_uf_od_coco-person.tflite) | Int8 | 416x416x3 | 64.80 % |
101
 
102
  \* EVAL_IOU = 0.5, NMS_THRESH = 0.5, SCORE_THRESH = 0.001, MAX_DETECTIONS = 100
103
 
 
65
 
66
 
67
  ### Reference **NPU** memory footprint based on COCO Person dataset (see Accuracy for details on dataset)
68
+ | Model | Dataset | Format | Resolution | Series | Internal RAM (KiB) | External RAM (KiB) | Weights Flash (KiB) | STEdgeAI Core version |
69
+ |---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-------------|----------|--------------|----------|----------------------|----------------------|-----------------------|-------------------------|
70
+ | [YOLOv8n per channel](https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/stedgeai_models/object_detection/yolov8n_192_quant_pc_uf_od_coco-person.tflite) | COCO-Person | Int8 | 192x192x3 | STM32N6 | 270 | 0 | 2936.52 | 3.0.0 |
71
+ | [YOLOv8n per channel](https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/stedgeai_models/object_detection/yolov8n_256_quant_pc_uf_od_coco-person.tflite) | COCO-Person | Int8 | 256x256x3 | STM32N6 | 624 | 0 | 2959.08 | 3.0.0 |
72
+ | [YOLOv8n per channel](https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/stedgeai_models/object_detection/yolov8n_320_quant_pc_uf_od_coco-person.tflite) | COCO-Person | Int8 | 320x320x3 | STM32N6 | 758.31 | 0 | 2963.51 | 3.0.0 |
73
+ | [YOLOv8n per channel](https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/stedgeai_models/object_detection/yolov8n_416_quant_pc_uf_od_coco-person.tflite) | COCO-Person | Int8 | 416x416x3 | STM32N6 | 2067.88 | 0 | 2972.00 | 3.0.0 |
74
+
75
  ### Reference **NPU** inference time based on COCO Person dataset (see Accuracy for details on dataset)
76
+ | Model | Dataset | Format | Resolution | Board | Execution Engine | Inference time (ms) | Inf / sec | STEdgeAI Core version |
77
+ |---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-------------|----------|--------------|---------------|--------------------|-----------------------|-------------|-------------------------|
78
+ | [YOLOv8n per channel](https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/stedgeai_models/object_detection/yolov8n_192_quant_pc_uf_od_coco-person.tflite) | COCO-Person | Int8 | 192x192x3 | STM32N6570-DK | NPU/MCU | 16.25 | 61.54 | 3.0.0 |
79
+ | [YOLOv8n per channel](https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/stedgeai_models/object_detection/yolov8n_256_quant_pc_uf_od_coco-person.tflite) | COCO-Person | Int8 | 256x256x3 | STM32N6570-DK | NPU/MCU | 23.84 | 41.96 | 3.0.0 |
80
+ | [YOLOv8n per channel](https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/stedgeai_models/object_detection/yolov8n_320_quant_pc_uf_od_coco-person.tflite) | COCO-Person | Int8 | 320x320x3 | STM32N6570-DK | NPU/MCU | 32.26 | 31.00 | 3.0.0 |
81
+ | [YOLOv8n per channel](https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/stedgeai_models/object_detection/yolov8n_416_quant_pc_uf_od_coco-person.tflite) | COCO-Person | Int8 | 416x416x3 | STM32N6570-DK | NPU/MCU | 54.64 | 18.30 | 3.0.0 |
82
+
83
+
84
+ ### Reference **MCU** memory footprint based on COCO Person dataset (see Accuracy for details on dataset)
85
+
86
+ | Model | Format | Resolution | Series | Activation RAM (KiB) | Runtime RAM (KiB) | Weights Flash (KiB) | Code Flash (KiB) | Total RAM (KiB) | Total Flash (KiB) | STEdgeAI Core version |
87
+ |---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|----------|--------------|----------|------------------------|---------------------|-----------------------|--------------------|--------------------|----------------------|-------------------------|
88
+ | [YOLOv8n per channel](https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/stedgeai_models/object_detection/yolov8n_192_quant_pc_uf_od_coco-person.tflite) | Int8 | 192x192x3 | STM32H7 | 367.69 | 62.11 | 2955.64 | 151.16 | 429.80 | 3106.80 | 3.0.0 |
89
+ | [YOLOv8n per channel](https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/stedgeai_models/object_detection/yolov8n_256_quant_pc_uf_od_coco-person.tflite) | Int8 | 256x256x3 | STM32H7 | 653.67 | 62.11 | 2959.08 | 151.29 | 715.78 | 3110.37 | 3.0.0 |
90
+ | [YOLOv8n per channel](https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/stedgeai_models/object_detection/yolov8n_320_quant_pc_uf_od_coco-person.tflite) | Int8 | 320x320x3 | STM32H7 | 1021.36 | 62.11 | 2963.51 | 151.41 | 1083.47 | 3114.92 | 3.0.0 |
91
+ | [YOLOv8n per channel](https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/stedgeai_models/object_detection/yolov8n_416_quant_pc_uf_od_coco-person.tflite) | Int8 | 416x416x3 | STM32H7 | 1726.09 | 62.11 | 2972.00 | 151.33 | 1788.20 | 3123.33 | 3.0.0 |
92
+
93
+
94
+ ### Reference **MCU** inference time based on COCO Person dataset (see Accuracy for details on dataset)
95
+
96
+ | Model | Format | Resolution | Board | Execution Engine | Frequency | Inference time (ms) | STEdgeAI Core version |
97
+ |---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|----------|--------------|------------------|--------------------|-------------|-----------------------|------------------------|
98
+ | [YOLOv8n per channel](https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/stedgeai_models/object_detection/yolov8n_192_quant_pc_uf_od_coco-person.tflite) | Int8 | 192x192x3 | STM32H747I-DISCO | 1 CPU | 400 MHz | 1612.46 | 3.0.0 |
99
+ | [YOLOv8n per channel](https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/stedgeai_models/object_detection/yolov8n_256_quant_pc_uf_od_coco-person.tflite) | Int8 | 256x256x3 | STM32H747I-DISCO | 1 CPU | 400 MHz | 2839.27 | 3.0.0 |
100
+ | [YOLOv8n per channel](https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/stedgeai_models/object_detection/yolov8n_320_quant_pc_uf_od_coco-person.tflite) | Int8 | 320x320x3 | STM32H747I-DISCO | 1 CPU | 400 MHz | 4442.59 | 3.0.0 |
101
+ | [YOLOv8n per channel](https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/stedgeai_models/object_detection/yolov8n_416_quant_pc_uf_od_coco-person.tflite) | Int8 | 416x416x3 | STM32H747I-DISCO | 1 CPU | 400 MHz | 7868.62 | 3.0.0 |
102
+
103
  ### Reference **MPU** inference time based on COCO Person dataset (see Accuracy for details on dataset)
104
  Model | Format | Resolution | Quantization | Board | Execution Engine | Frequency | Inference time (ms) | %NPU | %GPU | %CPU | X-LINUX-AI version | Framework |
105
  |-----------|--------|------------|---------------|-------------------|------------------|-----------|---------------------|-------|-------|------|--------------------|-----------------------|
 
116
 
117
  | Model | Format | Resolution | AP* |
118
  |-------|--------|------------|----------------|
119
+ | [YOLOv8n per channel](https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/stedgeai_models/object_detection/yolov8n_192_quant_pc_uf_od_coco-person.tflite) | Int8 | 192x192x3 | 53.58 % |
120
  | [YOLOv8n per channel](https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/stedgeai_models/object_detection/yolov8n_256_quant_pc_uf_od_coco-person.tflite) | Int8 | 256x256x3 | 58.40 % |
121
+ | [YOLOv8n per channel](https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/stedgeai_models/object_detection/yolov8n_320_quant_pc_uf_od_coco-person.tflite) | Int8 | 320x320x3 | 61.86 % |
122
+ | [YOLOv8n per channel](https://github.com/stm32-hotspot/ultralytics/blob/main/examples/YOLOv8-STEdgeAI/stedgeai_models/object_detection/yolov8n_416_quant_pc_uf_od_coco-person.tflite) | Int8 | 416x416x3 | 64.79 % |
123
 
124
  \* EVAL_IOU = 0.5, NMS_THRESH = 0.5, SCORE_THRESH = 0.001, MAX_DETECTIONS = 100
125