无需人脸检测，即可实时，6自由度3维人脸姿态估计方法 | 代码刚开源

import sys
sys.path.append('../../')
import numpy as np
import torch
from torchvision import transforms
from matplotlib import pyplot as plt
from tqdm.notebook import tqdm
from PIL import Image, ImageOps
import matplotlib.patches as patches
from scipy.spatial.transform import Rotation
import pandas as pd
from scipy.spatial import distance
import time
import os
import math
import scipy.io as sio
import random
from utils.renderer import Renderer
from utils.image_operations import expand_bbox_rectangle
from utils.pose_operations import get_pose
from img2pose import img2poseModel
from model_loader import load_model
from data_loader_lmdb import LMDBDataLoader
from dataclasses import dataclass

np.set_printoptions(suppress=True)

torch.random.manual_seed(42)
np.random.seed(42)

@dataclass
class Config:
    batch_size: int
    pin_memory: bool
    workers: int
    pose_mean: np.array
    pose_stddev: np.array
    noise_augmentation: bool
    contrast_augmentation: bool
    threed_68_points: str
    distributed: bool
renderer = Renderer(
    vertices_path="pose_references/vertices_trans.npy", 
    triangles_path="pose_references/triangles.npy"
)

threed_points = np.load('pose_references/reference_3d_68_points_trans.npy')

transform = transforms.Compose([transforms.ToTensor()])

# bounding box customization
# how much to expand in the width
BBOX_X_FACTOR = 1.1
# how much to expand in height
BBOX_Y_FACTOR = 1.1
# how much to expand the forehead
EXPAND_FOREHEAD = 0.3

DEPTH = 18
MAX_SIZE = 1400
MIN_SIZE = 600

POSE_MEAN = "models/WIDER_train_pose_mean_v1.npy"
POSE_STDDEV = "models/WIDER_train_pose_stddev_v1.npy"
MODEL_PATH = "models/img2pose_v1.pth"

pose_mean = np.load(POSE_MEAN)
pose_stddev = np.load(POSE_STDDEV)

img2pose_model = img2poseModel(
    DEPTH, MIN_SIZE, MAX_SIZE, 
    pose_mean=pose_mean, pose_stddev=pose_stddev,
    threed_68_points=threed_points,    
    bbox_x_factor=BBOX_X_FACTOR,
    bbox_y_factor=BBOX_Y_FACTOR,
    expand_forehead=EXPAND_FOREHEAD,
)
load_model(img2pose_model.fpn_model, MODEL_PATH, cpu_mode=str(img2pose_model.device) == "cpu", model_only=True)
img2pose_model.evaluate()


LMDB_FILE = "datasets/lmdb/WIDER_val_annotations.lmdb"

# load images that were already compressed from json list to lmdb file
lmdb_data_loader = LMDBDataLoader(
    Config(
        batch_size=1,
        pin_memory=True,
        workers=1,
        pose_mean=pose_mean,
        pose_stddev=pose_stddev,
        noise_augmentation=False,
        contrast_augmentation=False,
        threed_68_points='pose_references/reference_3d_68_points_trans.npy',
        distributed=False
    ),
    LMDB_FILE,
    train=True,
)

threshold = 0.8
total_imgs = 20

data_iter = iter(lmdb_data_loader)

for j in tqdm(range(total_imgs)):
    torch_img, target = next(data_iter)
    target = target[0]
        
    bboxes = []
    scores = []
    poses = []
    
    img = torch_img[0]
    img = img.squeeze()
    img = transforms.ToPILImage()(img).convert("RGB")
    ori_img = img.copy()
    

    run_img = img.copy()

    w, h = img.size

    min_size = min(w, h)
    max_size = max(w, h)
    
    # run on the original image size
    img2pose_model.fpn_model.module.set_max_min_size(max_size, min_size)

    res = img2pose_model.predict([transform(run_img)])

    res = res[0]

    for i in range(len(res["scores"])):
        if res["scores"][i] > threshold:
            bboxes.append(res["boxes"].cpu().numpy()[i].astype('int'))
            scores.append(res["scores"].cpu().numpy()[i].astype('float'))
            poses.append(res["dofs"].cpu().numpy()[i].astype('float'))
                
    (w, h) = img.size
    image_intrinsics = np.array([[w + h, 0, w // 2], [0, w + h, h // 2], [0, 0, 1]])
    
    plt.figure(figsize=(16, 16))    
    
    poses = np.asarray(poses)
    bboxes = np.asarray(bboxes)
    scores = np.asarray(scores)
    print(poses)
    if np.ndim(bboxes) == 1 and len(bboxes) > 0:
        bboxes = bboxes[np.newaxis, :]
        poses = poses[np.newaxis, :]        
        
    if len(bboxes) != 0:
        ranked = np.argsort(poses[:, 5])[::-1]
        poses = poses[ranked]
        bboxes = bboxes[ranked]
        scores = scores[ranked]

        for i in range(len(scores)):
            if scores[i] > threshold:
                bbox = bboxes[i]

                pose_pred = poses[i]
                pose_pred = np.asarray(pose_pred.squeeze())        

                trans_vertices = renderer.transform_vertices(img, [pose_pred])
                img = renderer.render(img, trans_vertices, alpha=1)  
                plt.gca().add_patch(patches.Rectangle((bbox[0], bbox[1]), bbox[2] - bbox[0], bbox[3] - bbox[1],linewidth=3,edgecolor='b',facecolor='none'))            
                img = Image.fromarray(img)

        plt.imshow(img)        
        plt.show() 

下载1：何恺明顶会分享
在「AI算法与图像处理」公众号后台回复：何恺明，即可下载。总共有6份PDF，涉及 ResNet、Mask RCNN等经典工作的总结分析
下载2：leetcode 开源书
在「AI算法与图像处理」公众号后台回复：leetcode，即可下载。每题都 runtime beats 100% 的开源好书，你值得拥有！

下载3 CVPR2020
在「AI算法与图像处理」公众号后台回复：CVPR2020，即可下载1467篇CVPR 2020论文个人微信（如果没有备注不拉群！）请注明：地区+学校/企业+研究方向+昵称

觉得不错就点亮在看吧