yatman
/
Image_Gen_Server
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

update + test

This commit is contained in:
zaqxs123456 2024-10-14 14:50:56 +08:00
parent e7b27a58f1
commit 4ea0b0280b
7 changed files with 367 additions and 44 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

72
app.py
View File

@ -1,6 +1,8 @@
import base64
import hashlib
import json
import random
import uuid
import websocket
from flask import Flask, request, jsonify
import sys
import openpose_gen as opg
@ -9,30 +11,80 @@ sys.path.append('./')
app = Flask(__name__)
@app.route('/coordinates', methods=['POST'])
def receive_coordinates():
@app.route('/gen_image', methods=['POST'])
def gen_image():
if request.is_json:
data = request.get_json()
coordinates = data['coordinates']
canvas_size = data['canvas_size']
pid = data['pid']
if not coordinates or not canvas_size:
return jsonify({"status": "error", "message": "Missing data"}), 422
opg.save_bodypose(canvas_size, canvas_size, coordinates)
openpose_image_path = opg.save_bodypose(canvas_size[0], canvas_size[1], coordinates, pid)
# gen_fencer_prompt(openpose_image_path, pid, opg.server_address)
return jsonify({"status": "success", "message": "Data received"}), 201
else:
return jsonify({"status": "error", "message": "Request must be JSON"}), 415
def open_websocket_connection():
server_address='127.0.0.1:8188'
client_id=str(uuid.uuid4())
ws = websocket.WebSocket()
ws.connect("ws://{}/ws?clientId={}".format(server_address, client_id))
return ws, server_address, client_id
@app.route('/gen_group_pic', methods=['POST'])
def init_gen_group_pic():
if request.is_json:
data = request.get_json()
coordinates_list = data['coordinates_list']
canvas_size = data['canvas_size']
pid = data['pid']
base_image = base64.b64decode(data['base_image'])
# save base image to ./test.png
with open("test.png", "wb") as f:
f.write(base_image)
if not coordinates_list or not canvas_size or not base_image or not pid:
return jsonify({"status": "error", "message": "Missing data"}), 422
for i in range(len(coordinates_list)):
coordinates_list[i] = coordinates_list[i]['coordinates']
openpose_image_path = opg.save_bodypose_mulit(canvas_size[0], canvas_size[1], coordinates_list, pid)
# gen_group_pic(openpose_image_path, base_image, pid, opg.server_address)
return jsonify({"status": "success", "message": "Data received"}), 201
else:
return jsonify({"status": "error", "message": "Request must be JSON"}), 415
def gen_fencer_prompt(openpose_image_path, pid, comfyUI_address):
with open("fencerAPI.json", "r") as f:
prompt_json = f.read()
prompt = json.loads(prompt_json)
openpose_image_name = opg.upload_image_circular_queue(openpose_image_path, 20, pid)
opg.upload_image("ref_black.png", "ref_black.png")
prompt["3"]["inputs"]["seed"] = random.randint(0, 10000000000)
prompt["29"]["inputs"]['image'] = "ref_black.png"
prompt["17"]["inputs"]['image'] = openpose_image_name
opg.queue_prompt(prompt, comfyUI_address)
def gen_group_pic(openpose_image_path, base_image, pid, comfyUI_address):
with open("groupAPI.json", "r") as f:
prompt_json = f.read()
prompt = json.loads(prompt_json)
openpose_image_name = opg.upload_image_circular_queue(openpose_image_path, 30, pid)
base_image_name = opg.upload_image_circular_queue(base_image, 30, pid)
prompt["3"]["inputs"]["seed"] = random.randint(0, 10000000000)
prompt["17"]["inputs"]['image'] = openpose_image_name
prompt["17"]["inputs"]['image'] = base_image_name
opg.queue_prompt(prompt, comfyUI_address)
if __name__ == '__main__':
app.run(debug=True)

BIN
body_pose_output_multi0000.png Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 15 KiB

215
fencerAPI copy.json Normal file
View File

@ -0,0 +1,215 @@
{
"3": {
"inputs": {
"seed": 688190695340079,
"steps": 3,
"cfg": 2,
"sampler_name": "dpmpp_sde",
"scheduler": "karras",
"denoise": 1,
"model": [
"32",
0
],
"positive": [
"22",
0
],
"negative": [
"22",
1
],
"latent_image": [
"5",
0
]
},
"class_type": "KSampler",
"_meta": {
"title": "KSampler"
}
},
"4": {
"inputs": {
"ckpt_name": "dreamshaperXL_sfwLightningDPMSDE.safetensors"
},
"class_type": "CheckpointLoaderSimple",
"_meta": {
"title": "Load Checkpoint"
}
},
"5": {
"inputs": {
"width": 1024,
"height": 1024,
"batch_size": 1
},
"class_type": "EmptyLatentImage",
"_meta": {
"title": "Empty Latent Image"
}
},
"6": {
"inputs": {
"text": "A fencer in full gear, fencing sword, 1 human, empty background, dark background, dark, empty",
"clip": [
"4",
1
]
},
"class_type": "CLIPTextEncode",
"_meta": {
"title": "CLIP Text Encode (Positive)"
}
},
"8": {
"inputs": {
"samples": [
"3",
0
],
"vae": [
"4",
2
]
},
"class_type": "VAEDecode",
"_meta": {
"title": "VAE Decode"
}
},
"9": {
"inputs": {
"filename_prefix": "Result",
"images": [
"8",
0
]
},
"class_type": "SaveImage",
"_meta": {
"title": "Save Image"
}
},
"17": {
"inputs": {
"image": "dance_01.png",
"upload": "image"
},
"class_type": "LoadImage",
"_meta": {
"title": "Load Image"
}
},
"19": {
"inputs": {
"control_net_name": "diffusion_pytorch_model.safetensors"
},
"class_type": "ControlNetLoader",
"_meta": {
"title": "Load ControlNet Model"
}
},
"22": {
"inputs": {
"strength": 0.9500000000000001,
"start_percent": 0,
"end_percent": 1,
"positive": [
"6",
0
],
"negative": [
"40",
0
],
"control_net": [
"19",
0
],
"image": [
"17",
0
],
"vae": [
"4",
2
]
},
"class_type": "ControlNetApplyAdvanced",
"_meta": {
"title": "Apply ControlNet"
}
},
"28": {
"inputs": {
"ipadapter_file": "ip-adapter-plus_sdxl_vit-h.safetensors"
},
"class_type": "IPAdapterModelLoader",
"_meta": {
"title": "IPAdapter Model Loader"
}
},
"29": {
"inputs": {
"image": "ref_black.png",
"upload": "image"
},
"class_type": "LoadImage",
"_meta": {
"title": "Load Image"
}
},
"31": {
"inputs": {
"clip_name": "CLIP-ViT-H-14-laion2B-s32B-b79K.safetensors"
},
"class_type": "CLIPVisionLoader",
"_meta": {
"title": "Load CLIP Vision"
}
},
"32": {
"inputs": {
"weight": 1,
"weight_type": "style and composition",
"combine_embeds": "add",
"start_at": 0,
"end_at": 1,
"embeds_scaling": "V only",
"model": [
"4",
0
],
"ipadapter": [
"28",
0
],
"image": [
"29",
0
],
"clip_vision": [
"31",
0
]
},
"class_type": "IPAdapterAdvanced",
"_meta": {
"title": "IPAdapter Advanced"
}
},
"40": {
"inputs": {
"text": "blurry, drawing, horror, distorted, malformed, naked, cartoon, anime, out of focus, dull, muted colors, boring pose, no action, distracting background, colorful, (face:5.0), bad hand, bad anatomy, worst quality, ai generated images, low quality, average quality, smoke, background, three arms, three hands, white light, (light:5.0), (shadow:5.0), (floor:5.0)\n\nembedding:ac_neg1, embedding:ac_neg2, embedding:badhandv4, embedding:DeepNegative_xl_v1, embedding:NEGATIVE_HANDS, embedding:negativeXL_D, embedding:unaestheticXL_cbp62 -neg, embedding:verybadimagenegative_v1.3, embedding:ziprealism_neg, ",
"clip": [
"4",
1
]
},
"class_type": "CLIPTextEncode",
"_meta": {
"title": "CLIP Text Encode (Negative)"
}
}
}

2
fencer_02_no_ipa.json
View File

@ -51,7 +51,7 @@
},
"6": {
"inputs": {
"text": "A fencer in full gear, fencing sword, 1 human, empty background, dark background, dark, empty, 1 sword, sword in hand",
"text": "costume party, trick-or-treating, candy apples, pumpkin carving, haunted hayride, spooky stories, dress-up, decorations, Halloween parade, haunted corn maze, bobbing for apples, ghost stories, scary movies, pumpkin patch, fall festival, Halloween games, costume contest",
"clip": [
"4",
1

30
fov_cal.py Normal file
View File

@ -0,0 +1,30 @@
import math
def hfov_to_vfov(hFOV, aspect_ratio):
return 2 * math.degrees(math.atan(math.tan(math.radians(hFOV / 2)) / aspect_ratio))
def hfov_to_dfov(hFOV, aspect_ratio):
return 2 * math.degrees(math.atan(math.sqrt(1 + (1 / aspect_ratio)**2) * math.tan(math.radians(hFOV / 2))))
def vfov_to_hfov(vFOV, aspect_ratio):
return 2 * math.degrees(math.atan(aspect_ratio * math.tan(math.radians(vFOV / 2))))
def vfov_to_dfov(vFOV, aspect_ratio):
return 2 * math.degrees(math.atan(math.sqrt(1 + aspect_ratio**2) * math.tan(math.radians(vFOV / 2))))
def dfov_to_hfov(dFOV, aspect_ratio):
return 2 * math.degrees(math.atan(math.tan(math.radians(dFOV / 2)) / math.sqrt(1 + (1 / aspect_ratio)**2)))
def dfov_to_vfov(dFOV, aspect_ratio):
return 2 * math.degrees(math.atan(math.tan(math.radians(dFOV / 2)) / math.sqrt(1 + aspect_ratio**2)))
fov = 60
aspect_ratio = 2/1
print("h2v:", hfov_to_vfov(fov, aspect_ratio))
print("h2d:", hfov_to_dfov(fov, aspect_ratio))
print("v2h:", vfov_to_hfov(fov, aspect_ratio))
print("v2d:", vfov_to_dfov(fov, aspect_ratio))
print("d2h:", dfov_to_hfov(fov, aspect_ratio))
print("d2v:", dfov_to_vfov(fov, aspect_ratio))

92
openpose_gen.py
View File

@ -11,6 +11,7 @@ import urllib
from urllib import request, parse
from requests_toolbelt.multipart.encoder import MultipartEncoder
import sys
import hashlib
sys.path.append('./')
def is_normalized(keypoints: List[skel.Keypoint]) -> bool:
@ -87,7 +88,7 @@ def coordinates_to_keypoints(coordinates: list) -> List[skel.Keypoint]:
keypoints = [skel.Keypoint(coordinates[i], coordinates[i + 1]) for i in range(0, len(coordinates), 3)]
return keypoints
def save_bodypose(width: int, height: int, coordinates: list):
def save_bodypose(width: int, height: int, coordinates: list, pid: str) -> None:
if not hasattr(save_bodypose, 'counter'):
save_bodypose.counter = 0 # Initialize the counter attribute
@ -96,51 +97,76 @@ def save_bodypose(width: int, height: int, coordinates: list):
canvas = draw_bodypose(canvas, keypoints, skel.coco_limbSeq, skel.coco_colors)
# Save as body_pose_output0000.png, body_pose_output0001.png, ...
image_name = 'output/body_pose_output%04d.png' % save_bodypose.counter
cv2.imwrite(image_name, canvas)
gen_image(image_name)
image_path = 'output/body_pose_output%04d.png' % save_bodypose.counter
cv2.imwrite(image_path, canvas)
save_bodypose.counter += 1 # Increment the counter
server_address = "http://127.0.0.1:8188"
return image_path
def queue_prompt(prompt):
def save_bodypose_mulit(width: int, height: int, coordinates_list: list, pid: str) -> None:
if not hasattr(save_bodypose_mulit, 'counter'):
save_bodypose_mulit.counter = 0 # Initialize the counter attribute
canvas = np.zeros((height, width, 3), dtype=np.uint8)
for coordinates in coordinates_list:
keypoints = coordinates_to_keypoints(coordinates)
canvas = draw_bodypose(canvas, keypoints, skel.coco_limbSeq, skel.coco_colors)
# Save as body_pose_output0000.png, body_pose_output0001.png, ...
image_path = 'output/body_pose_output_multi%04d.png' % save_bodypose_mulit.counter
cv2.imwrite(image_path, canvas)
save_bodypose_mulit.counter += 1 # Increment the counter
return image_path
server_address = "localhost:8188"
def queue_prompt(prompt, server_address):
p = {"prompt": prompt}
data = json.dumps(p).encode('utf-8')
req = request.Request(server_address + "/prompt", data=data)
req = request.Request("http://{}/prompt".format(server_address), data=data)
request.urlopen(req)
def upload_image(input_path, name, server_address, image_type="input", overwrite=False):
with open(input_path, 'rb') as file:
multipart_data = MultipartEncoder(
fields= {
'image': (name, file, 'image/png'),
'type': image_type,
'overwrite': str(overwrite).lower()
}
)
def upload_image(input_image, name, server_address, image_type="input", overwrite=False):
# Check if input_image is a valid file path
if isinstance(input_image, str) and os.path.isfile(input_image):
file = open(input_image, 'rb')
close_file = True
else:
file = input_image
close_file = False
data = multipart_data
headers = { 'Content-Type': multipart_data.content_type }
request = urllib.request.Request("http://{}/upload/image".format(server_address), data=data, headers=headers)
with urllib.request.urlopen(request) as response:
return response.read()
try:
multipart_data = MultipartEncoder(
fields={
'image': (name, file, 'image/png'),
'type': image_type,
'overwrite': str(overwrite).lower()
}
)
def gen_image(openpose_image_path):
data = multipart_data
headers = {'Content-Type': multipart_data.content_type}
request = urllib.request.Request("http://{}/upload/image".format(server_address), data=data, headers=headers)
with urllib.request.urlopen(request) as response:
return response.read()
finally:
if close_file:
file.close()
# read fencerAPI.json into prompt_text
with open("fencerAPI.json", "r") as f:
prompt_json = f.read()
prompt = json.loads(prompt_json)
def upload_image_circular_queue(image_path, size, unqiue_id, server_address):
# create a dict in this function to store the counter for each unique_id, key is the unique_id, value is the counter
if not hasattr(upload_image_circular_queue, 'id_counter_dict'):
upload_image_circular_queue.id_counter_dict = {}
# upload images
upload_image("ref_black.png", "ref_black.png", server_address)
upload_image(openpose_image_path, "openpose_image.png", server_address)
if unqiue_id not in upload_image_circular_queue.id_counter_dict:
upload_image_circular_queue.id_counter_dict[unqiue_id] = 0
prompt["3"]["inputs"]["seed"] = random.randint(0, 10000000000)
prompt["29"]["inputs"]['image'] = "ref_black.png"
prompt["17"]["inputs"]['image'] = "openpose_image.png"
image_name = hashlib.sha256((unqiue_id + str(upload_image_circular_queue.id_counter_dict[unqiue_id])).encode('utf-8')).hexdigest() + ".png"
upload_image_circular_queue.id_counter_dict[unqiue_id] += 1 % size
upload_image(image_path, image_name, server_address)
queue_prompt(prompt)
return image_name
def main():
directory = './fixed'

BIN
test.png Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 1.5 MiB