MediaPipe-Pose-Estimation: Optimized for Mobile Deployment
Detect and track human face, hand, and torso in real-time images and video streams
The MediaPipe Pose Landmark Detector is a machine learning pipeline that predicts bounding boxes and pose skeletons of the face, hands, and torso in an image.
This model is an implementation of MediaPipe-Pose-Estimation found here.
This repository provides scripts to run MediaPipe-Pose-Estimation on Qualcomm® devices. More details on model performance across various devices, can be found here.
Model Details
- Model Type: Model_use_case.pose_estimation
- Model Stats:
- Input resolution: 256x256
- Number of parameters (PoseDetector): 815K
- Model size (PoseDetector) (float): 3.14 MB
- Number of parameters (PoseLandmarkDetector): 3.36M
- Model size (PoseLandmarkDetector) (float): 12.9 MB
| Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model |
|---|---|---|---|---|---|---|---|---|
| PoseDetector | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 5.743 ms | 0 - 22 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseDetector | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 5.73 ms | 0 - 21 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseDetector | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 2.164 ms | 0 - 30 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseDetector | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 2.181 ms | 0 - 30 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseDetector | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 0.9 ms | 0 - 29 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseDetector | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 0.854 ms | 0 - 19 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseDetector | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 1.095 ms | 0 - 18 MB | NPU | MediaPipe-Pose-Estimation.onnx.zip |
| PoseDetector | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 1.671 ms | 0 - 23 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseDetector | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 1.618 ms | 0 - 21 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseDetector | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 5.743 ms | 0 - 22 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseDetector | float | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 5.73 ms | 0 - 21 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseDetector | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 0.895 ms | 0 - 28 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseDetector | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 0.853 ms | 0 - 18 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseDetector | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 2.641 ms | 0 - 26 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseDetector | float | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 2.528 ms | 0 - 26 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseDetector | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 0.904 ms | 0 - 5 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseDetector | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 0.846 ms | 0 - 18 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseDetector | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 1.671 ms | 0 - 23 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseDetector | float | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 1.618 ms | 0 - 21 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseDetector | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 0.648 ms | 0 - 41 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseDetector | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 0.615 ms | 0 - 32 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseDetector | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 0.777 ms | 0 - 37 MB | NPU | MediaPipe-Pose-Estimation.onnx.zip |
| PoseDetector | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 0.529 ms | 0 - 30 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseDetector | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 0.508 ms | 0 - 29 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseDetector | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 0.65 ms | 0 - 23 MB | NPU | MediaPipe-Pose-Estimation.onnx.zip |
| PoseDetector | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | TFLITE | 0.405 ms | 0 - 24 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseDetector | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | QNN_DLC | 0.399 ms | 0 - 25 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseDetector | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | ONNX | 0.571 ms | 0 - 29 MB | NPU | MediaPipe-Pose-Estimation.onnx.zip |
| PoseDetector | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 1.017 ms | 7 - 7 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseDetector | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 1.043 ms | 3 - 3 MB | NPU | MediaPipe-Pose-Estimation.onnx.zip |
| PoseLandmarkDetector | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 3.189 ms | 0 - 35 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseLandmarkDetector | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 3.161 ms | 1 - 28 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseLandmarkDetector | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 1.009 ms | 0 - 52 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseLandmarkDetector | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 1.165 ms | 1 - 47 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseLandmarkDetector | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 0.824 ms | 0 - 78 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseLandmarkDetector | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 0.808 ms | 0 - 46 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseLandmarkDetector | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 1.146 ms | 0 - 44 MB | NPU | MediaPipe-Pose-Estimation.onnx.zip |
| PoseLandmarkDetector | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 1.459 ms | 0 - 34 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseLandmarkDetector | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 1.441 ms | 0 - 27 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseLandmarkDetector | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 3.189 ms | 0 - 35 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseLandmarkDetector | float | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 3.161 ms | 1 - 28 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseLandmarkDetector | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 0.802 ms | 0 - 80 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseLandmarkDetector | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 0.787 ms | 0 - 45 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseLandmarkDetector | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 1.442 ms | 0 - 36 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseLandmarkDetector | float | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 1.41 ms | 1 - 29 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseLandmarkDetector | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 0.803 ms | 0 - 80 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseLandmarkDetector | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 0.804 ms | 0 - 45 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseLandmarkDetector | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 1.459 ms | 0 - 34 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseLandmarkDetector | float | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 1.441 ms | 0 - 27 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseLandmarkDetector | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 0.604 ms | 0 - 56 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseLandmarkDetector | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 0.598 ms | 1 - 53 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseLandmarkDetector | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 0.785 ms | 0 - 54 MB | NPU | MediaPipe-Pose-Estimation.onnx.zip |
| PoseLandmarkDetector | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 0.471 ms | 0 - 42 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseLandmarkDetector | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 0.494 ms | 1 - 35 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseLandmarkDetector | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 0.705 ms | 0 - 33 MB | NPU | MediaPipe-Pose-Estimation.onnx.zip |
| PoseLandmarkDetector | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | TFLITE | 0.411 ms | 0 - 34 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseLandmarkDetector | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | QNN_DLC | 0.425 ms | 1 - 34 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseLandmarkDetector | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | ONNX | 0.663 ms | 1 - 34 MB | NPU | MediaPipe-Pose-Estimation.onnx.zip |
| PoseLandmarkDetector | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 1.073 ms | 32 - 32 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseLandmarkDetector | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 1.163 ms | 7 - 7 MB | NPU | MediaPipe-Pose-Estimation.onnx.zip |
| PoseDetector | w8a8 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 1.074 ms | 0 - 21 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseDetector | w8a8 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 1.0 ms | 0 - 22 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseDetector | w8a8 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 0.552 ms | 0 - 32 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseDetector | w8a8 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 0.572 ms | 0 - 29 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseDetector | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 0.336 ms | 0 - 11 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseDetector | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 0.319 ms | 0 - 22 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseDetector | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 0.56 ms | 0 - 23 MB | NPU | MediaPipe-Pose-Estimation.onnx.zip |
| PoseDetector | w8a8 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 0.554 ms | 0 - 22 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseDetector | w8a8 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 1.874 ms | 0 - 22 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseDetector | w8a8 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | TFLITE | 1.697 ms | 0 - 25 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseDetector | w8a8 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | QNN_DLC | 1.949 ms | 0 - 27 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseDetector | w8a8 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | ONNX | 10.52 ms | 8 - 20 MB | CPU | MediaPipe-Pose-Estimation.onnx.zip |
| PoseDetector | w8a8 | RB5 (Proxy) | Qualcomm® QCS8250 (Proxy) | TFLITE | 6.632 ms | 1 - 4 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseDetector | w8a8 | RB5 (Proxy) | Qualcomm® QCS8250 (Proxy) | ONNX | 8.563 ms | 9 - 12 MB | CPU | MediaPipe-Pose-Estimation.onnx.zip |
| PoseDetector | w8a8 | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 1.074 ms | 0 - 21 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseDetector | w8a8 | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 1.0 ms | 0 - 22 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseDetector | w8a8 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 0.335 ms | 0 - 12 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseDetector | w8a8 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 0.319 ms | 0 - 22 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseDetector | w8a8 | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 0.962 ms | 0 - 27 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseDetector | w8a8 | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 0.914 ms | 0 - 27 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseDetector | w8a8 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 0.335 ms | 0 - 12 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseDetector | w8a8 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 0.314 ms | 0 - 6 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseDetector | w8a8 | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 0.554 ms | 0 - 22 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseDetector | w8a8 | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 1.874 ms | 0 - 22 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseDetector | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 0.237 ms | 0 - 34 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseDetector | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 0.213 ms | 0 - 38 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseDetector | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 0.354 ms | 0 - 38 MB | NPU | MediaPipe-Pose-Estimation.onnx.zip |
| PoseDetector | w8a8 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 0.201 ms | 0 - 23 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseDetector | w8a8 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 0.187 ms | 0 - 25 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseDetector | w8a8 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 0.297 ms | 0 - 27 MB | NPU | MediaPipe-Pose-Estimation.onnx.zip |
| PoseDetector | w8a8 | Snapdragon 7 Gen 4 QRD | Snapdragon® 7 Gen 4 Mobile | TFLITE | 0.432 ms | 0 - 23 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseDetector | w8a8 | Snapdragon 7 Gen 4 QRD | Snapdragon® 7 Gen 4 Mobile | QNN_DLC | 0.403 ms | 0 - 28 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseDetector | w8a8 | Snapdragon 7 Gen 4 QRD | Snapdragon® 7 Gen 4 Mobile | ONNX | 9.429 ms | 10 - 26 MB | CPU | MediaPipe-Pose-Estimation.onnx.zip |
| PoseDetector | w8a8 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | TFLITE | 0.167 ms | 0 - 23 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseDetector | w8a8 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | QNN_DLC | 0.161 ms | 0 - 28 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseDetector | w8a8 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | ONNX | 0.29 ms | 0 - 25 MB | NPU | MediaPipe-Pose-Estimation.onnx.zip |
| PoseDetector | w8a8 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 0.423 ms | 14 - 14 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseDetector | w8a8 | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 0.467 ms | 0 - 0 MB | NPU | MediaPipe-Pose-Estimation.onnx.zip |
| PoseLandmarkDetector | w8a8 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 0.878 ms | 0 - 28 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseLandmarkDetector | w8a8 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 0.801 ms | 0 - 30 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseLandmarkDetector | w8a8 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 0.421 ms | 0 - 37 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseLandmarkDetector | w8a8 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 0.448 ms | 0 - 35 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseLandmarkDetector | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 0.339 ms | 0 - 34 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseLandmarkDetector | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 0.302 ms | 0 - 32 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseLandmarkDetector | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 0.624 ms | 0 - 33 MB | NPU | MediaPipe-Pose-Estimation.onnx.zip |
| PoseLandmarkDetector | w8a8 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 2.037 ms | 0 - 28 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseLandmarkDetector | w8a8 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 0.526 ms | 0 - 29 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseLandmarkDetector | w8a8 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | TFLITE | 0.945 ms | 0 - 37 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseLandmarkDetector | w8a8 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | QNN_DLC | 1.075 ms | 0 - 37 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseLandmarkDetector | w8a8 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | ONNX | 9.099 ms | 7 - 26 MB | CPU | MediaPipe-Pose-Estimation.onnx.zip |
| PoseLandmarkDetector | w8a8 | RB5 (Proxy) | Qualcomm® QCS8250 (Proxy) | TFLITE | 7.017 ms | 0 - 2 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseLandmarkDetector | w8a8 | RB5 (Proxy) | Qualcomm® QCS8250 (Proxy) | ONNX | 7.791 ms | 6 - 12 MB | CPU | MediaPipe-Pose-Estimation.onnx.zip |
| PoseLandmarkDetector | w8a8 | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 0.878 ms | 0 - 28 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseLandmarkDetector | w8a8 | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 0.801 ms | 0 - 30 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseLandmarkDetector | w8a8 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 0.338 ms | 0 - 32 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseLandmarkDetector | w8a8 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 0.307 ms | 0 - 31 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseLandmarkDetector | w8a8 | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 0.747 ms | 0 - 34 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseLandmarkDetector | w8a8 | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 0.683 ms | 0 - 35 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseLandmarkDetector | w8a8 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 0.344 ms | 0 - 33 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseLandmarkDetector | w8a8 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 0.31 ms | 0 - 32 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseLandmarkDetector | w8a8 | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 2.037 ms | 0 - 28 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseLandmarkDetector | w8a8 | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 0.526 ms | 0 - 29 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseLandmarkDetector | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 0.231 ms | 0 - 38 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseLandmarkDetector | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 0.209 ms | 0 - 39 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseLandmarkDetector | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 0.363 ms | 0 - 44 MB | NPU | MediaPipe-Pose-Estimation.onnx.zip |
| PoseLandmarkDetector | w8a8 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 0.2 ms | 0 - 27 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseLandmarkDetector | w8a8 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 0.187 ms | 0 - 38 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseLandmarkDetector | w8a8 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 0.322 ms | 0 - 36 MB | NPU | MediaPipe-Pose-Estimation.onnx.zip |
| PoseLandmarkDetector | w8a8 | Snapdragon 7 Gen 4 QRD | Snapdragon® 7 Gen 4 Mobile | TFLITE | 0.35 ms | 0 - 34 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseLandmarkDetector | w8a8 | Snapdragon 7 Gen 4 QRD | Snapdragon® 7 Gen 4 Mobile | QNN_DLC | 0.314 ms | 0 - 35 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseLandmarkDetector | w8a8 | Snapdragon 7 Gen 4 QRD | Snapdragon® 7 Gen 4 Mobile | ONNX | 8.52 ms | 9 - 29 MB | CPU | MediaPipe-Pose-Estimation.onnx.zip |
| PoseLandmarkDetector | w8a8 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | TFLITE | 0.193 ms | 0 - 35 MB | NPU | MediaPipe-Pose-Estimation.tflite |
| PoseLandmarkDetector | w8a8 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | QNN_DLC | 0.168 ms | 0 - 33 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseLandmarkDetector | w8a8 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | ONNX | 0.33 ms | 0 - 38 MB | NPU | MediaPipe-Pose-Estimation.onnx.zip |
| PoseLandmarkDetector | w8a8 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 0.433 ms | 26 - 26 MB | NPU | MediaPipe-Pose-Estimation.dlc |
| PoseLandmarkDetector | w8a8 | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 0.477 ms | 3 - 3 MB | NPU | MediaPipe-Pose-Estimation.onnx.zip |
Installation
Install the package via pip:
pip install qai-hub-models
Configure Qualcomm® AI Hub Workbench to run this model on a cloud-hosted device
Sign-in to Qualcomm® AI Hub Workbench with your
Qualcomm® ID. Once signed in navigate to Account -> Settings -> API Token.
With this API token, you can configure your client to run models on the cloud hosted devices.
qai-hub configure --api_token API_TOKEN
Navigate to docs for more information.
Demo off target
The package contains a simple end-to-end demo that downloads pre-trained weights and runs this model on a sample input.
python -m qai_hub_models.models.mediapipe_pose.demo
The above demo runs a reference implementation of pre-processing, model inference, and post processing.
NOTE: If you want running in a Jupyter Notebook or Google Colab like environment, please add the following to your cell (instead of the above).
%run -m qai_hub_models.models.mediapipe_pose.demo
Run model on a cloud-hosted device
In addition to the demo, you can also run the model on a cloud-hosted Qualcomm® device. This script does the following:
- Performance check on-device on a cloud-hosted device
- Downloads compiled assets that can be deployed on-device for Android.
- Accuracy check between PyTorch and on-device outputs.
python -m qai_hub_models.models.mediapipe_pose.export
How does this work?
This export script leverages Qualcomm® AI Hub to optimize, validate, and deploy this model on-device. Lets go through each step below in detail:
Step 1: Compile model for on-device deployment
To compile a PyTorch model for on-device deployment, we first trace the model
in memory using the jit.trace and then call the submit_compile_job API.
import torch
import qai_hub as hub
from qai_hub_models.models.mediapipe_pose import Model
# Load the model
torch_model = Model.from_pretrained()
# Device
device = hub.Device("Samsung Galaxy S25")
# Trace model
input_shape = torch_model.get_input_spec()
sample_inputs = torch_model.sample_inputs()
pt_model = torch.jit.trace(torch_model, [torch.tensor(data[0]) for _, data in sample_inputs.items()])
# Compile model on a specific device
compile_job = hub.submit_compile_job(
model=pt_model,
device=device,
input_specs=torch_model.get_input_spec(),
)
# Get target model to run on-device
target_model = compile_job.get_target_model()
Step 2: Performance profiling on cloud-hosted device
After compiling models from step 1. Models can be profiled model on-device using the
target_model. Note that this scripts runs the model on a device automatically
provisioned in the cloud. Once the job is submitted, you can navigate to a
provided job URL to view a variety of on-device performance metrics.
profile_job = hub.submit_profile_job(
model=target_model,
device=device,
)
Step 3: Verify on-device accuracy
To verify the accuracy of the model on-device, you can run on-device inference on sample input data on the same cloud hosted device.
input_data = torch_model.sample_inputs()
inference_job = hub.submit_inference_job(
model=target_model,
device=device,
inputs=input_data,
)
on_device_output = inference_job.download_output_data()
With the output of the model, you can compute like PSNR, relative errors or spot check the output with expected output.
Note: This on-device profiling and inference requires access to Qualcomm® AI Hub Workbench. Sign up for access.
Deploying compiled model to Android
The models can be deployed using multiple runtimes:
TensorFlow Lite (
.tfliteexport): This tutorial provides a guide to deploy the .tflite model in an Android application.QNN (
.soexport ): This sample app provides instructions on how to use the.soshared library in an Android application.
View on Qualcomm® AI Hub
Get more details on MediaPipe-Pose-Estimation's performance across various devices here. Explore all available models on Qualcomm® AI Hub
License
- The license for the original implementation of MediaPipe-Pose-Estimation can be found here.
- The license for the compiled assets for on-device deployment can be found here
References
Community
- Join our AI Hub Slack community to collaborate, post questions and learn more about on-device AI.
- For questions or feedback please reach out to us.
- Downloads last month
- 823
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
🙋
Ask for provider support