Jimmy

@protossw512 I encounter the same problem as yours! The ratio between inter-pupil and inter-occular is around 0.71 as I estimated! Whats more this paper meets the ratio！ see http://openaccess.thecvf.com/content_ECCV_2018/papers/Roberto_Valle_A_Deeply-initialized_Coarse-to-fine_ECCV_2018_paper.pdf

> To reproduce: > > import torch > from torchvision import models > from aimet_torch.quantsim import QuantizationSimModel > > BATCH_SIZE = 1 > input_shape = (BATCH_SIZE, 3, 224, 224) >...

> In your face profiling code, in the main.m, vertex_full, tri_full and vertexm_full are loaded from .mat file, The question is how I can generate these array by myself? Have...

> this is my sample code > > def make_bin(path,save_name,shift_num,image_size): > > ``` > even_list, odd_list, truths = even_odd_split(path,shift_num)#replace your own test dataset code > new_dataset = [] > >...