human-pose-estimation.pytorch
human-pose-estimation.pytorch copied to clipboard
Code for exporting native model to ONNX and visualizing network work
Converting to ONNX
To convert PyTorch model to ONNX format run script:
python pose_estimation/convert.py --cfg experiments/coco/resnet152/384x288_d256x3_adam_lr1e-3.yaml --model-file pose_estimation/pose_resnet_152_384x288.pth.tar
Visualization
To show processed image run script:
python pose_estimation\demo_picture.py --img pose_estimation\test\hugh_laurie.jpg --model .\models\onnx\pose_resnet_152_384x288.onnx --type ONNX --width 656 --height 384
Result

Script processing image via opencv dnn module