3d-object-detection.pytorch

3d-object-detection.pytorch

Input

(Image from Objectron Dataset https://github.com/google-research-datasets/Objectron/blob/master/notebooks/Download%20Data.ipynb)

Detection model shape : (1, 3, H, W)
Regression model shape : (1, 3, 224, 224)

Output

Detection model shape : (N, 5), (N,)
Regression model shape : (9, 1, 9, 2), (1, 9)

Category

OBJECTRON_CLASSES = [
    'bike', 'book', 'bottle', 'cereal_box', 'camera', 
    'chair', 'cup', 'laptop', 'shoe'
]

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 3d-object-detection.pytorch.py

If you want to specify the input image, put the image path after the --input option.
You can use --savepath option to change the name of the output file to save.

$ python3 3d-object-detection.pytorch.py --input IMAGE_PATH --savepath SAVE_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 3d-object-detection.pytorch.py --video VIDEO_PATH

Reference

• 3d-object-detection.pytorch
• Objectron Dataset

Framework

Pytorch

Model Format

ONNX opset=11

Netron

mnv2_ssd_300_2_heads.onnx.prototxt
regression_model_epoch120.onnx.prototxt