xsl
443cfa298f
包含： - hair_service_sd: 主服务（换发型/换发色/生发，端口8801） - photo_service: LoRA调度+训练（端口32678） - hair_grow_service: 调试测试页（端口8888，含4个测试页） - 批量训练脚本（batch_train_hairstyles.py） - 发际线mask自动识别（hairline_mask.py，4种方案） - 手绘mask换发型（hair_swap_manual.py） - 文档：README.md + LARGE_FILES.md + docs/ 大文件（模型权重200G、训练数据123G）已排除，见 LARGE_FILES.md OSS/COS密钥已脱敏为环境变量，原文件备份在本地
273 lines
12 KiB
Python
273 lines
12 KiB
Python
# -*- coding: utf-8 -*-
|
||
"""换发型（带可视化中间产物）
|
||
|
||
和换发型 /api/swapHair/v1 完全一致的工作流，但保存每一步的中间产物
|
||
用于前端可视化展示换发型过程。
|
||
|
||
mask 处理与换发型一致：origin_matting - 刘海圆 ∪ new_matting
|
||
（与生发不同：不减刘海、不加手绘mask）
|
||
"""
|
||
import os
|
||
import json
|
||
import glob
|
||
import time
|
||
import shutil
|
||
import base64
|
||
|
||
import cv2
|
||
import numpy as np
|
||
|
||
from gen_super_image import webui_img2img
|
||
from common.logger import config
|
||
from utils.landmark_processor import high_quality_warpAffine
|
||
from utils.landmark_processor import pts_1k_to_137
|
||
|
||
|
||
def hair_swap_viz(origin_img, hair_id, hairstyle_process, landmark_processor,
|
||
task_id, is_hr=True, strict_mask=False):
|
||
"""换发型 + 保存中间产物。
|
||
|
||
返回:
|
||
result_img: 最终结果图 BGR
|
||
steps: [{title, desc, images:[{label, b64}]}] 每步的中间产物
|
||
"""
|
||
start_all = time.time()
|
||
steps = []
|
||
|
||
hairstyle_dir = config.get('default', 'hairstyleDir')
|
||
user_dir = config.get('default', 'userDir')
|
||
train_dir = config.get('default', 'train_dir')
|
||
userInfo_dir = config.get('default', 'userInfo_dir')
|
||
|
||
hair_material_dir = os.path.join(train_dir, hair_id)
|
||
material_save_path = os.path.join(hairstyle_dir, hair_id)
|
||
ref_img_path = os.path.join(material_save_path, "ref_rgb_8uc3_768.png")
|
||
if not os.path.exists(ref_img_path):
|
||
raise FileNotFoundError(f"发型材质不存在: {ref_img_path}")
|
||
|
||
user_img_name = f"swapviz_{task_id}.jpg"
|
||
new_user_img_path = os.path.join(user_dir, user_img_name)
|
||
if os.path.exists(new_user_img_path):
|
||
os.remove(new_user_img_path)
|
||
cv2.imwrite(new_user_img_path, origin_img)
|
||
origin_img = cv2.imread(new_user_img_path)
|
||
|
||
ref_img = cv2.imread(ref_img_path)
|
||
|
||
# === 步骤1: 输入图 ===
|
||
steps.append({
|
||
"title": "① 输入",
|
||
"desc": "用户人像 + 目标发型参考图",
|
||
"images": [
|
||
{"label": "用户原图", "b64": _enc(origin_img)},
|
||
{"label": f"发型参考图({hair_id})", "b64": _enc(cv2.resize(ref_img, (origin_img.shape[1], origin_img.shape[0])))},
|
||
]
|
||
})
|
||
|
||
# === 步骤2: 粗推理 infer_hairstyle_diy_jy ===
|
||
start = time.time()
|
||
work_dir = os.path.join(userInfo_dir, task_id)
|
||
os.makedirs(work_dir, exist_ok=True)
|
||
import torch
|
||
with torch.no_grad():
|
||
img_res, status, _, landmarks_1k, isEyeOccluded = hairstyle_process.infer_hairstyle_diy_jy(
|
||
origin_img, ref_img, os.path.join(work_dir, task_id), f"{task_id}.png")
|
||
if status != 0:
|
||
raise RuntimeError("发型粗推理失败")
|
||
print(f"[swap_viz] 粗推理: {time.time()-start:.1f}s")
|
||
|
||
user_material_dir = os.path.join(work_dir, task_id)
|
||
hair_matting_path = os.path.join(user_material_dir, "hair_mask_2.png")
|
||
user_orig_mask_path = os.path.join(user_material_dir, "user_orig_mask.png")
|
||
origin_matting = cv2.imread(user_orig_mask_path, cv2.IMREAD_GRAYSCALE)
|
||
new_matting = cv2.imread(hair_matting_path, cv2.IMREAD_GRAYSCALE)
|
||
|
||
steps.append({
|
||
"title": "② 粗推理",
|
||
"desc": f"把目标发型粗略合成到用户脸上（耗时{time.time()-start:.1f}s）",
|
||
"images": [
|
||
{"label": "粗推理结果", "b64": _enc(img_res)},
|
||
]
|
||
})
|
||
|
||
# === 步骤3: mask 合并（换发型：减刘海 ∪ 新发型）===
|
||
start = time.time()
|
||
# 刘海圆（与换发型功能8一致）
|
||
no_bang_result = origin_matting.copy()
|
||
try:
|
||
landmarks_137 = pts_1k_to_137(np.asarray(landmarks_1k))
|
||
# 眼睛中点（137点第7、14点）
|
||
eye_centers = []
|
||
for i in [7, 14]:
|
||
if i < len(landmarks_137):
|
||
eye_centers.append(landmarks_137[i])
|
||
if len(eye_centers) == 2:
|
||
cx = int((eye_centers[0][0] + eye_centers[1][0]) / 2)
|
||
cy = int((eye_centers[0][1] + eye_centers[1][1]) / 2)
|
||
r = int(np.linalg.norm(np.array(eye_centers[0]) - np.array(eye_centers[1])))
|
||
cv2.circle(no_bang_result, (cx, cy), r, 0, -1)
|
||
except Exception as e:
|
||
print(f"[swap_viz] 刘海圆计算失败({e})，不减刘海")
|
||
|
||
matting_merge = np.max(np.stack([no_bang_result, new_matting], axis=2), axis=2).astype(np.uint8)
|
||
|
||
steps.append({
|
||
"title": "③ 生成重绘mask",
|
||
"desc": "用户原头发mask(减刘海圆) ∪ 新发型mask = 需要重绘的区域",
|
||
"images": [
|
||
{"label": "用户原头发mask", "b64": _enc_mask(origin_matting)},
|
||
{"label": "新发型mask", "b64": _enc_mask(new_matting)},
|
||
{"label": "合并mask(重绘区)", "b64": _enc_mask(matting_merge)},
|
||
]
|
||
})
|
||
|
||
# === 步骤4: warpAffine 变换 ===
|
||
dst_size = (1152, 1536) if is_hr else (576, 768)
|
||
box_info = hairstyle_process.get_body_info(img_res)
|
||
box_w = box_info[2] - box_info[0]
|
||
box_h = box_info[3] - box_info[1]
|
||
scale = min(dst_size[1] / max(box_h, 1), dst_size[0] / max(box_w, 1))
|
||
rotate_center = [(box_info[2] + box_info[0]) * 0.5, (box_info[3] + box_info[1]) * 0.5]
|
||
M = cv2.getRotationMatrix2D(rotate_center, 0, scale)
|
||
M[:, 2] += np.float32([dst_size[0] * 0.5, dst_size[1] * 0.5]) - np.float32(rotate_center)
|
||
|
||
crop_result = high_quality_warpAffine(img_res, M, dst_size)
|
||
crop_matting = cv2.warpAffine(matting_merge, M, dst_size)
|
||
mask = (crop_matting > 10).astype(np.float32)
|
||
dilate_kernel = (6, 18) if is_hr else (3, 9)
|
||
mask_dilate = cv2.dilate(mask, np.ones(dilate_kernel, np.uint8))
|
||
mask_dilate = np.clip(mask_dilate * 255, 0, 255).astype(np.uint8)
|
||
final_img = crop_result
|
||
|
||
steps.append({
|
||
"title": "④ warpAffine变换",
|
||
"desc": f"把图和mask对齐裁剪到{dst_size[0]}x{dst_size[1]}送入webui",
|
||
"images": [
|
||
{"label": "变换后的图", "b64": _enc(cv2.resize(crop_result, (crop_result.shape[1]//2, crop_result.shape[0]//2)))},
|
||
{"label": "变换后的mask", "b64": _enc_mask(cv2.resize(mask_dilate, (mask_dilate.shape[1]//2, mask_dilate.shape[0]//2), interpolation=cv2.INTER_NEAREST))},
|
||
]
|
||
})
|
||
|
||
# === 步骤5: 读取 config/prompt ===
|
||
config_json_path = os.path.join(material_save_path, "config.json")
|
||
with open(config_json_path, "r") as f:
|
||
in_gender = json.load(f)["gender"]
|
||
images_dir = os.path.join(hair_material_dir, "images")
|
||
txt_dir = os.path.join(images_dir, os.listdir(images_dir)[0])
|
||
txt_path = glob.glob(txt_dir + '/*.txt')[0]
|
||
with open(txt_path, 'r') as f:
|
||
p_tag = f.readline()
|
||
if "titor hairstyle, faceless, no human, gray background, simple background" in p_tag:
|
||
p_tag = p_tag[p_tag.find("simple background, ") + len("simple background, "):]
|
||
else:
|
||
p_tag = ""
|
||
|
||
# === 步骤6: webui 推理（走 photo_service + LoRA）===
|
||
start = time.time()
|
||
sd_result = webui_img2img(
|
||
img=final_img, mask_img=mask_dilate,
|
||
in_gender=in_gender, task_id=task_id,
|
||
hair_id=hair_id, lora_material_path=hair_material_dir,
|
||
tag=p_tag, is_hr=is_hr,
|
||
denoising_strength=0.6, inference_port="57860")
|
||
print(f"[swap_viz] webui推理: {time.time()-start:.1f}s")
|
||
|
||
steps.append({
|
||
"title": "⑤ SD推理(LoRA)",
|
||
"desc": f"webui img2img inpainting，加载发型LoRA，在mask区重绘新发型（耗时{time.time()-start:.1f}s）",
|
||
"images": [
|
||
{"label": "webui输出", "b64": _enc(cv2.resize(sd_result, (sd_result.shape[1]//2, sd_result.shape[0]//2)))},
|
||
]
|
||
})
|
||
|
||
# === 步骤7: warpAffine 贴回原图 ===
|
||
M_inv = cv2.invertAffineTransform(M)
|
||
# 先算整框贴回（原方式：整个裁剪框覆盖）
|
||
result_full = origin_img.copy()
|
||
cv2.warpAffine(sd_result, M_inv, (origin_img.shape[1], origin_img.shape[0]),
|
||
dst=result_full, borderMode=cv2.BORDER_TRANSPARENT, flags=cv2.INTER_LANCZOS4)
|
||
|
||
# 严格按mask贴回：只覆盖mask_dilate逆变换后的区域
|
||
result_strict = origin_img.copy()
|
||
sd_result_back = np.zeros_like(origin_img)
|
||
cv2.warpAffine(sd_result, M_inv, (origin_img.shape[1], origin_img.shape[0]),
|
||
dst=sd_result_back, borderMode=cv2.BORDER_CONSTANT, borderValue=(0,0,0), flags=cv2.INTER_LANCZOS4)
|
||
# mask_dilate逆变换到原图尺寸
|
||
mask_back = np.zeros(origin_img.shape[:2], dtype=np.uint8)
|
||
cv2.warpAffine(mask_dilate, M_inv, (origin_img.shape[1], origin_img.shape[0]),
|
||
dst=mask_back, borderMode=cv2.BORDER_CONSTANT, borderValue=0, flags=cv2.INTER_NEAREST)
|
||
# 轻微膨胀+高斯模糊让边缘自然
|
||
mask_back = cv2.dilate(mask_back, np.ones((5, 5), np.uint8))
|
||
mask_blend = (mask_back.astype(np.float32) / 255)[..., None]
|
||
result_strict = (sd_result_back.astype(np.float32) * mask_blend +
|
||
origin_img.astype(np.float32) * (1 - mask_blend))
|
||
result_strict = np.clip(result_strict, 0, 255).astype(np.uint8)
|
||
# mask_back=0 的区域（裁剪框外）恢复原图
|
||
result_strict[mask_back == 0] = origin_img[mask_back == 0]
|
||
|
||
# 根据strict_mask选择返回哪个
|
||
result = result_strict if strict_mask else result_full
|
||
|
||
# === 步骤7.5: 消除接缝（泊松无缝融合）===
|
||
# strict_mask模式下mask边缘有接缝，用seamlessClone消除
|
||
# 原理：保持sd_result内部梯度，把边缘梯度过渡到原图，消除色差/亮度突变
|
||
if strict_mask:
|
||
start_enhance = time.time()
|
||
try:
|
||
# sd_result逆变换到原图尺寸
|
||
sd_back = np.zeros_like(origin_img)
|
||
cv2.warpAffine(sd_result, M_inv, (origin_img.shape[1], origin_img.shape[0]),
|
||
dst=sd_back, borderMode=cv2.BORDER_CONSTANT, borderValue=(0,0,0), flags=cv2.INTER_LANCZOS4)
|
||
# mask逆变换 + 膨胀（扩大融合区域）
|
||
mask_back2 = np.zeros(origin_img.shape[:2], dtype=np.uint8)
|
||
cv2.warpAffine(mask_dilate, M_inv, (origin_img.shape[1], origin_img.shape[0]),
|
||
dst=mask_back2, borderMode=cv2.BORDER_CONSTANT, borderValue=0, flags=cv2.INTER_NEAREST)
|
||
mask_back2 = cv2.dilate(mask_back2, np.ones((9, 9), np.uint8))
|
||
# 泊松融合需要mask外接矩形中心
|
||
ys2, xs2 = np.where(mask_back2 > 10)
|
||
if len(ys2) > 10:
|
||
cx2 = int((xs2.min() + xs2.max()) / 2)
|
||
cy2 = int((ys2.min() + ys2.max()) / 2)
|
||
# NORMAL_CLONE: 在mask区域用sd_back的颜色，边缘梯度自然过渡
|
||
result = cv2.seamlessClone(sd_back, origin_img, mask_back2, (cx2, cy2), cv2.NORMAL_CLONE)
|
||
print(f"[swap_viz] 泊松融合消除接缝: {time.time()-start_enhance:.1f}s")
|
||
except Exception as e:
|
||
print(f"[swap_viz] 泊松融合失败({e})，回退到严格mask结果")
|
||
|
||
steps.append({
|
||
"title": "⑥ 贴回+增强重绘" + ("（严格按mask+消除接缝）" if strict_mask else "（整框覆盖+增强）"),
|
||
"desc": "先贴回原图，再用低denoise(0.35)全图增强重绘消除mask接缝（耗时见日志）",
|
||
"images": [
|
||
{"label": "用户原图", "b64": _enc(origin_img)},
|
||
{"label": "整框覆盖(增强前)", "b64": _enc(result_full)},
|
||
{"label": "严格mask(增强前)", "b64": _enc(result_strict)},
|
||
{"label": "最终(增强后)", "b64": _enc(result)},
|
||
]
|
||
})
|
||
|
||
# 清理临时文件
|
||
try:
|
||
shutil.rmtree(user_material_dir)
|
||
os.remove(new_user_img_path)
|
||
except Exception:
|
||
pass
|
||
|
||
print(f"[swap_viz] 总耗时: {time.time()-start_all:.1f}s")
|
||
return result, steps
|
||
|
||
|
||
def _enc(img):
|
||
"""ndarray -> base64 jpeg"""
|
||
_, buf = cv2.imencode(".jpg", img, [cv2.IMWRITE_JPEG_QUALITY, 80])
|
||
return base64.b64encode(buf).decode()
|
||
|
||
|
||
def _enc_mask(mask):
|
||
"""单通道mask -> base64 jpeg（伪彩色化便于查看）"""
|
||
if mask.ndim == 2:
|
||
colored = cv2.applyColorMap(mask, cv2.COLORMAP_JET)
|
||
else:
|
||
colored = mask
|
||
_, buf = cv2.imencode(".jpg", colored, [cv2.IMWRITE_JPEG_QUALITY, 80])
|
||
return base64.b64encode(buf).decode()
|