qualcomm
/

MediaPipe-Pose-Estimation

Keypoint Detection

PyTorch

TF Lite

real_time

android

Model card Files Files and versions Community

qaihm-bot commited on Jul 16

Commit

7ddcc69

•

1 Parent(s): 1b40cf3

Upload README.md with huggingface_hub

Browse files

Files changed (1) hide show

README.md +12 -12

README.md CHANGED Viewed

@@ -36,10 +36,10 @@ More details on model performance across various devices, can be found
 | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
 | ---|---|---|---|---|---|---|---|
-| Samsung Galaxy S23 Ultra (Android 13) | Snapdragon® 8 Gen 2 | TFLite | 0.826 ms | 0 - 1 MB | FP16 | NPU |  [MediaPipePoseDetector.tflite](https://huggingface.co/qualcomm/MediaPipe-Pose-Estimation/blob/main/MediaPipePoseDetector.tflite)
-| Samsung Galaxy S23 Ultra (Android 13) | Snapdragon® 8 Gen 2 | TFLite | 1.229 ms | 0 - 3 MB | FP16 | NPU |  [MediaPipePoseLandmarkDetector.tflite](https://huggingface.co/qualcomm/MediaPipe-Pose-Estimation/blob/main/MediaPipePoseLandmarkDetector.tflite)
-| Samsung Galaxy S23 Ultra (Android 13) | Snapdragon® 8 Gen 2 | QNN Model Library | 0.879 ms | 0 - 4 MB | FP16 | NPU |  [MediaPipePoseDetector.so](https://huggingface.co/qualcomm/MediaPipe-Pose-Estimation/blob/main/MediaPipePoseDetector.so)
-| Samsung Galaxy S23 Ultra (Android 13) | Snapdragon® 8 Gen 2 | QNN Model Library | 1.34 ms | 0 - 13 MB | FP16 | NPU |  [MediaPipePoseLandmarkDetector.so](https://huggingface.co/qualcomm/MediaPipe-Pose-Estimation/blob/main/MediaPipePoseLandmarkDetector.so)
@@ -100,17 +100,17 @@ python -m qai_hub_models.models.mediapipe_pose.export
 ```
 Profile Job summary of MediaPipePoseDetector
 --------------------------------------------------
-Device: Snapdragon X Elite CRD (11)
-Estimated Inference Time: 1.00 ms
-Estimated Peak Memory Range: 0.50-0.50 MB
-Compute Units: NPU (139) | Total (139)
 Profile Job summary of MediaPipePoseLandmarkDetector
 --------------------------------------------------
-Device: Snapdragon X Elite CRD (11)
-Estimated Inference Time: 1.43 ms
-Estimated Peak Memory Range: 0.75-0.75 MB
-Compute Units: NPU (305) | Total (305)
 ```

 | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
 | ---|---|---|---|---|---|---|---|
+| Samsung Galaxy S23 Ultra (Android 13) | Snapdragon® 8 Gen 2 | TFLite | 0.837 ms | 0 - 2 MB | FP16 | NPU |  [MediaPipePoseDetector.tflite](https://huggingface.co/qualcomm/MediaPipe-Pose-Estimation/blob/main/MediaPipePoseDetector.tflite)
+| Samsung Galaxy S23 Ultra (Android 13) | Snapdragon® 8 Gen 2 | TFLite | 1.011 ms | 0 - 2 MB | FP16 | NPU |  [MediaPipePoseLandmarkDetector.tflite](https://huggingface.co/qualcomm/MediaPipe-Pose-Estimation/blob/main/MediaPipePoseLandmarkDetector.tflite)
+| Samsung Galaxy S23 Ultra (Android 13) | Snapdragon® 8 Gen 2 | QNN Model Library | 0.884 ms | 0 - 15 MB | FP16 | NPU |  [MediaPipePoseDetector.so](https://huggingface.co/qualcomm/MediaPipe-Pose-Estimation/blob/main/MediaPipePoseDetector.so)
+| Samsung Galaxy S23 Ultra (Android 13) | Snapdragon® 8 Gen 2 | QNN Model Library | 1.083 ms | 0 - 16 MB | FP16 | NPU |  [MediaPipePoseLandmarkDetector.so](https://huggingface.co/qualcomm/MediaPipe-Pose-Estimation/blob/main/MediaPipePoseLandmarkDetector.so)
 ```
 Profile Job summary of MediaPipePoseDetector
 --------------------------------------------------
+Device: SA8255 (Proxy) (13)
+Estimated Inference Time: 0.89 ms
+Estimated Peak Memory Range: 0.02-5.94 MB
+Compute Units: NPU (140) | Total (140)
 Profile Job summary of MediaPipePoseLandmarkDetector
 --------------------------------------------------
+Device: SA8255 (Proxy) (13)
+Estimated Inference Time: 1.08 ms
+Estimated Peak Memory Range: 0.02-11.66 MB
+Compute Units: NPU (292) | Total (292)
 ```