Profile Job Results
Jobs
j5mq9oqdp
Results Ready
Name
mediapipe_face_FaceLandmarkDetector
Target Device
- QCS6490 (Proxy)
- Android 12
- Qualcomm® QCS6490
Creator
ai-hub-support@qti.qualcomm.com
Target Model
Input Specs
image: uint8[1, 192, 192, 3]Completion Time
5/20/2025, 10:50:56 PM
Versions
- TensorFlow Lite: 2.17.0
- QAIRT: v2.32.6.250402152434_116405
- QNN TfLite Delegate: v2.32.6.250402152434_116405
- Android: 12 (SP1A.210812.016)
- AI Hub: aihub-2025.05.14.0
Estimated Inference Time
374 μs
Estimated Peak Memory Usage
0 ‑ 17 MB
Compute Units
NPU
108
| Stage | Time | Memory |
|---|---|---|
First App Load | 475 ms | 29‑40 MB |
Subsequent App Load | 422 ms | 18‑41 MB |
Inference | 374 μs | 0‑17 MB |
| TensorFlow Lite | Value |
|---|---|
| number_of_threads | 4 |
| QNN Delegate | Value |
|---|---|
| backend_type | kHtpBackend |
| log_level | kLogLevelWarn |
| htp_options.performance_mode | kHtpBurst |
| htp_options.precision | kHtpFp16 |
| htp_options.useConvHmx | true |
Sign up to run this model on a hosted Qualcomm® device!
Run on device