PyRender for SMLP/-X/-H

需要包：

• smplx
• pyrender
• numpy
• trimesh
• opencv-python
• matplotlib
• torch

pip install -r requirements.txt

Usage：

将classes文件夹放入项目root目录（运行的根目录），需要调用处用法：

from classes.PyRenSmpl import PyRenSmpl
'''
参数说明：
smpl_model_path: path to smpl model
        save_path: path to save output result
        mv_width: meshviewer width      float
        mv_height: meshviewer height    float
        bg_color: background color     float(4)
        device: device

        camera_dict: dict of camera
            ca_rotate:camera rotate angle   float(3)[80, 0, 30]
            ca_translate: camera translate position   float(3)[1.2, -2.3, 0.5]
  
        light_dict: dict of light
            li_rotate:light rotate angle   float(3)[50, 0, 10]
            li_translate: light translate position   float(3)[0, 0, 1]
            li_intens:light intensity                   float  3.0

        render_ground:need a ground?        bool
        gr_translate: ground translate      float(3)
        gr_extent:ground size               float(3)
        gr_color:ground color               float
'''
#这些dict自己设置
camera_dict={
        'ca_rotate':[80, 0, 30],
        'ca_translate':[1.2, -2.3, 0.5],
    }
light_dict={
        'li_rotate':[50, 0, 10],
        'li_translate':[0, 0, 1],
        'li_intens': 3.0,
    }
input_dicts = {
        'device':device,
        'smpl_model_path':'./SMPL_models/',
        'save_path':cfg.result_dir + f'/{args.mode}',
        'mv_width':600,
        'mv_height':900,
        'bg_color':[1.0, 1.0, 1.0, 1.0],
        'camera_dict':camera_dict,
        'light_dict':light_dict,
        'render_ground': True,
        'gr_translate':[0, 0, -1.02],
        'gr_extent':[20,12,0.1],
        'gr_color':0.9,
    }
myrender = PyRenSmpl(input_dicts)
#调用
myrender.render_videos(
	sequence=sequence,
	key=key,
	smpl_model=smpl_model
)

大概：