LIP-JPPNet-TensorFlow
LIP-JPPNet-TensorFlow copied to clipboard
Help wanted
How can I generate a lip_val_set.csv file for my own dataset?
Hi @UmmaSaimaRahman , you can generate a lip_val_set.csv for your custom dataset following the script create_heatmaps.py
in the datasets/lip
folder. Hope that helps. Thanks.
Hi @minar09. Can you provide me a resource or tool for how to annotate the custom images according to README.md?
@UmmaSaimaRahman , for human parsing annotation, you can start with https://github.com/wkentaro/labelme or https://github.com/kyamagu/js-segment-annotator. For pose annotation tool: https://github.com/aimagelab/human-pose-annotation-tool Please note that I didn't use these tools myself, so I don't know how good are these. You can find better ones in Github or etc. or maybe use some popular pre-trained network models like OpenPose (https://github.com/CMU-Perceptual-Computing-Lab/openpose) or CIHP-PGN (https://github.com/Engineering-Course/CIHP_PGN) etc.