tucan9389 / PoseEstimation-MLKit

The example of running Pose Estimation using ML Kit
https://github.com/motlabs/iOS-Proejcts-with-ML-Models
MIT License
17 stars 2 forks source link

PoseEstimation-MLKit

platform-ios swift-version lisence

This project is Pose Estimation on iOS with ML Kit.
If you are interested in iOS + Machine Learning, visit here you can see various DEMOs.

Jointed Keypoints Concatenated heatmap
PoseEstimation-MLKit-hourglass.gif (preparing...)

How it works

how_it_works

Video source: https://www.youtube.com/watch?v=EM16LBKBEgI

Requirements

Download model

Get PoseEstimationForMobile's model

Pose Estimation model for TensorFlow Lite(model.tflite)
☞ Download TensorFlow Lite model model_cpm.tflite or hourglass.tflite.

input_name_shape_dict = {"image:0":[1,224,224,3]} image_input_names=["image:0"]
output_feature_names = ['Convolutional_Pose_Machine/stage_5_out:0']

-in https://github.com/edvardHua/PoseEstimationForMobile

Matadata

cpm hourglass
Input shape [1, 192, 192, 3] [1, 192, 192, 3]
Output shape [1, 96, 96, 14] [1, 48, 48, 14]
Input node name image image
Output node name Convolutional_Pose_Machine/stage_5_out hourglass_out_3

Inference Time

cpm hourglass
iPhone XS (TODO) (TODO)
iPhone XS Max (TODO) (TODO)
iPhone XR (TODO) (TODO)
iPhone X 57 ms 33 ms
iPhone 8+ (TODO) (TODO)
iPhone 8 (TODO) (TODO)
iPhone 7 (TODO) (TODO)
iPhone 6 (TODO) (TODO)

Get your own model

Or you can use your own PoseEstimation model

Build & Run

1. Prerequisites

1.1 Import pose estimation model

1.2 Add permission in info.plist for device's camera access

prerequest_001_plist

2. Dependencies

-

3. Code

(Preparing...)