search

mks0601 / I2L-MeshNet_RELEASE

Official PyTorch implementation of "I2L-MeshNet: Image-to-Lixel Prediction Network for Accurate 3D Human Pose and Mesh Estimation from a Single RGB Image", ECCV 2020
MIT License
720 stars 128 forks source link

The thickness of the generated mesh #125

Open Nenne524 opened 1 year ago

Nenne524 commented 1 year ago

Hello! When I run demo.py with an image I brought myself, the generated mesh has no thickness, what should I do? Please tell me. Screenshot from 2023-06-30 15-34-52

mks0601 commented 1 year ago

Could you use 'param' stage?

Nenne524 commented 1 year ago

Sorry for the late reply. I ran "python demo.py --gpu 0 --stage param --test_epoch 8". I followed the instructions in the quick demo. The bbox is already set up.