zaverichintan5
zaverichintan5
@cleardusk For 2D augmentation, are the 3DMM parameters calculated using HPEN code or just the pose matrix is changed?
Thanks for prompt reply. How is t3d changed for 2d augmented (rotated and translated) images? Is it divided by some factor or based some formula?
Is this the correct way of normalising the params: ``` all_params = _load(osp.join(d, 'params_600k.mat')) R = all_params['R'] t = all_params['t3d'] t = np.expand_dims(t, axis=1) R_t = np.concatenate([R, t], axis=1) R_t_flat...
Thanks for the quick reply. Is the keypoints available for the augmented images dataset? Or any link to the dataset used to train higher expressions model?
@cleardusk any updates on the pre-trained model with 60+29 or 199+29 dimensional parameters ?
Change the generate.m last 4 lines to this: ``` save -mat7-binary 'Out/BFM.mat' model save -mat7-binary 'Out/BFM_info.mat' model_info save -mat7-binary 'Out/BFM_UV.mat' UV ```
Stuck with Ubuntu 18.04 and python3.6. Any solutions or workaround ? Or anyone has some pointers?