Spaces:
Running
on
T4
Running
on
T4
import warnings | |
warnings.filterwarnings('ignore') | |
import subprocess, io, os, sys, time | |
from loguru import logger | |
# os.system("pip install diffuser==0.6.0") | |
# os.system("pip install transformers==4.29.1") | |
os.environ["CUDA_VISIBLE_DEVICES"] = "0" | |
if os.environ.get('IS_MY_DEBUG') is None: | |
result = subprocess.run(['pip', 'install', '-e', 'GroundingDINO'], check=True) | |
print(f'pip install GroundingDINO = {result}') | |
# result = subprocess.run(['pip', 'list'], check=True) | |
# print(f'pip list = {result}') | |
sys.path.insert(0, './GroundingDINO') | |
import gradio as gr | |
import argparse | |
import copy | |
import numpy as np | |
import torch | |
from PIL import Image, ImageDraw, ImageFont, ImageOps | |
# Grounding DINO | |
import GroundingDINO.groundingdino.datasets.transforms as T | |
from GroundingDINO.groundingdino.models import build_model | |
from GroundingDINO.groundingdino.util import box_ops | |
from GroundingDINO.groundingdino.util.slconfig import SLConfig | |
from GroundingDINO.groundingdino.util.utils import clean_state_dict, get_phrases_from_posmap | |
import cv2 | |
import numpy as np | |
import matplotlib.pyplot as plt | |
from lama_cleaner.model_manager import ModelManager | |
from lama_cleaner.schema import Config as lama_Config | |
# segment anything | |
from segment_anything import build_sam, SamPredictor, SamAutomaticMaskGenerator | |
# diffusers | |
import PIL | |
import requests | |
import torch | |
from io import BytesIO | |
from diffusers import StableDiffusionInpaintPipeline | |
from huggingface_hub import hf_hub_download | |
from utils import computer_info | |
# relate anything | |
from ram_utils import iou, sort_and_deduplicate, relation_classes, MLP, show_anns, ram_show_mask | |
from ram_train_eval import RamModel,RamPredictor | |
from mmengine.config import Config as mmengine_Config | |
from app import * | |
config_file = 'GroundingDINO/groundingdino/config/GroundingDINO_SwinT_OGC.py' | |
ckpt_repo_id = "ShilongLiu/GroundingDINO" | |
ckpt_filenmae = "groundingdino_swint_ogc.pth" | |
sam_checkpoint = './sam_vit_h_4b8939.pth' | |
output_dir = "outputs" | |
device = 'cpu' | |
os.makedirs(output_dir, exist_ok=True) | |
groundingdino_model = None | |
sam_device = None | |
sam_model = None | |
sam_predictor = None | |
sam_mask_generator = None | |
sd_pipe = None | |
lama_cleaner_model= None | |
ram_model = None | |
kosmos_model = None | |
kosmos_processor = None | |
def get_args(): | |
argparser = argparse.ArgumentParser() | |
argparser.add_argument("--input_image", "-i", type=str, default="", help="") | |
argparser.add_argument("--text", "-t", type=str, default="", help="") | |
argparser.add_argument("--output_image", "-o", type=str, default="", help="") | |
args = argparser.parse_args() | |
return args | |
# usage: | |
# python app_cli.py --input_image dog.png --text dog --output_image dog_remove.png | |
if __name__ == '__main__': | |
args = get_args() | |
logger.info(f'\nargs={args}\n') | |
logger.info(f'loading models ... ') | |
# set_device() # If you have enough GPUs, you can open this comment | |
groundingdino_model = load_groundingdino_model('cpu') | |
load_sam_model(device) | |
# load_sd_model(device) | |
load_lama_cleaner_model(device) | |
# load_ram_model(device) | |
input_image = Image.open(args.input_image) | |
output_images, _ = run_anything_task(input_image = input_image, | |
text_prompt = args.text, | |
task_type = 'remove', | |
inpaint_prompt = '', | |
box_threshold = 0.3, | |
text_threshold = 0.25, | |
iou_threshold = 0.8, | |
inpaint_mode = "merge", | |
mask_source_radio = "type what to detect below", | |
remove_mode = "rectangle", # ["segment", "rectangle"] | |
remove_mask_extend = "10", | |
num_relation = 5, | |
kosmos_input = None, | |
cleaner_size_limit = -1, | |
) | |
if len(output_images) > 0: | |
logger.info(f'save result to {args.output_image} ... ') | |
output_images[-1].save(args.output_image) | |
# count = 0 | |
# for output_image in output_images: | |
# count += 1 | |
# if isinstance(output_image, np.ndarray): | |
# output_image = PIL.Image.fromarray(output_image.astype(np.uint8)) | |
# output_image.save(args.output_image.replace(".", f"_{count}.")) | |