fashionai-key-points-detection

Cascaded Pyramid Network for FashionAI Key Points Detection

Input

blouse

dress

outwear

skirt

trousers

(Image from https://tianchi.aliyun.com/museum7/?spm=5176.14046517.J_9711814210.23.2bd17c0aFQzXFg#/newprodetail?productId=7)

Output

blouse

dress

outwear

skirt

trousers

Example keypoints

Example keypoints of the five clothing categories are as follows.

(Image from https://github.com/HiKapok/tf.fashionAI/blob/master/demos/outline.jpg)

Usage

Automatically downloads the onnx and prototxt files on the first run. It is necessary to be connected to the Internet while downloading.

For the sample image,

$ python3 fashionai-key-points-detection.py

You can specify the "clothing type" by specifying after the --clothing-type option. The clothing type is selected from blouse, dress, outwear.

$ python3 fashionai-key-points-detection.py --clothing-type blouse

If you want to specify the input image, put the image path after the --input option.

$ python3 fashionai-key-points-detection.py --input IMAGE_PATH

By adding the --video option, you can input the video.
If you pass 0 as an argument to VIDEO_PATH, you can use the webcam input instead of the video file.

$ python3 fashionai-key-points-detection.py --video VIDEO_PATH

Reference

• A Pytorch Implementation of Cascaded Pyramid Network for FashionAI Key Points Detection

Framework

Pytorch

Model Format

ONNX opset = 11

Netron

• blouse_100.onnx.prototxt
• dress_100.onnx.prototxt
• outwear_100.onnx.prototxt
• skirt_100.onnx.prototxt
• trousers_100.onnx.prototxt