<div align="center"> <h2> MetaSeg: Segment Anything 仓库的打包版本 </h2> <div> <img width="1000" alt="teaser" src="https://yellow-cdn.veclightyear.com/835a84d5/3225093a-17a0-42f2-9646-15e0652b933d.gif"> </div> <a href="https://pepy.tech/project/metaseg"><img src="https://pepy.tech/badge/metaseg" alt="downloads"></a> <a href="https://huggingface.co/spaces/ArtGAN/metaseg-webui"><img src="https://yellow-cdn.veclightyear.com/835a84d5/43354402-5314-4d20-9575-ca3788c55311.svg" alt="HuggingFace Spaces"></a> </div> <p align="center"> <a href="https://pypi.org/project/metaseg" target="_blank"> <img src="https://img.shields.io/pypi/v/metaseg?color=%2334D058&label=pypi%20package" alt="包版本"> </a> <a href="https://pypi.org/project/metaseg" target="_blank"> <img src="https://img.shields.io/pypi/dm/metaseg?color=red" alt="下载次数"> </a> <a href="https://pypi.org/project/metaseg" target="_blank"> <img src="https://yellow-cdn.veclightyear.com/835a84d5/80f2a4b2-166a-4ab9-8065-be04c233e335.svg?color=%2334D058" alt="支持的Python版本"> </a> <a href="https://pypi.org/project/metaseg" target="_blank"> <img src="https://img.shields.io/pypi/status/metaseg?color=orange" alt="项目状态"> </a> <a href="https://results.pre-commit.ci/latest/github/kadirnar/segment-anything-video/main" target="_blank"> <img src="https://yellow-cdn.veclightyear.com/835a84d5/5601b2d7-1da0-4b86-88a5-b73f1fa35194.svg" alt="pre-commit.ci"> </a> </p>

这个仓库是segment-anything模型的打包版本。

安装

pip install metaseg

使用方法

from metaseg import SegAutoMaskPredictor, SegManualMaskPredictor

# 如果GPU内存不足，请减少points_per_side和points_per_batch的值。

# 图像处理
results = SegAutoMaskPredictor().image_predict(
    source="image.jpg",
    model_type="vit_l", # vit_l, vit_h, vit_b
    points_per_side=16,
    points_per_batch=64,
    min_area=0,
    output_path="output.jpg",
    show=True,
    save=False,
)

# 视频处理
results = SegAutoMaskPredictor().video_predict(
    source="video.mp4",
    model_type="vit_l", # vit_l, vit_h, vit_b
    points_per_side=16,
    points_per_batch=64,
    min_area=1000,
    output_path="output.mp4",
)

# 手动框选和点选

# 图像处理
results = SegManualMaskPredictor().image_predict(
    source="image.jpg",
    model_type="vit_l", # vit_l, vit_h, vit_b
    input_point=[[100, 100], [200, 200]],
    input_label=[0, 1],
    input_box=[100, 100, 200, 200], # 或 [[100, 100, 200, 200], [100, 100, 200, 200]]
    multimask_output=False,
    random_color=False,
    show=True,
    save=False,
)

# 视频处理

results = SegManualMaskPredictor().video_predict(
    source="video.mp4",
    model_type="vit_l", # vit_l, vit_h, vit_b
    input_point=[0, 0, 100, 100],
    input_label=[0, 1],
    input_box=None,
    multimask_output=False,
    random_color=False,
    output_path="output.mp4",
)

SAHI + Segment Anything

pip install sahi metaseg

from metaseg.sahi_predict import SahiAutoSegmentation, sahi_sliced_predict

image_path = "image.jpg"
boxes = sahi_sliced_predict(
    image_path=image_path,
    detection_model_type="yolov5",  # yolov8, detectron2, mmdetection, torchvision
    detection_model_path="yolov5l6.pt",
    conf_th=0.25,
    image_size=1280,
    slice_height=256,
    slice_width=256,
    overlap_height_ratio=0.2,
    overlap_width_ratio=0.2,
)

SahiAutoSegmentation().image_predict(
    source=image_path,
    model_type="vit_b",
    input_box=boxes,
    multimask_output=False,
    random_color=False,
    show=True,
    save=False,
)

FalAI(云端GPU) + Segment Anything

pip install metaseg fal_serverless
fal-serverless auth login

# 自动蒙版
from metaseg import falai_automask_image

image = falai_automask_image(
    image_path="image.jpg",
    model_type="vit_b",
    points_per_side=16,
    points_per_batch=32,
    min_area=0,
)
image.show() # 显示图像
image.save("output.jpg") # 保存图像

# 手动蒙版
from metaseg import falai_manuelmask_image

image = falai_manualmask_image(
    image_path="image.jpg",
    model_type="vit_b",
    input_point=[[100, 100], [200, 200]],
    input_label=[0, 1],
    input_box=[100, 100, 200, 200], # 或 [[100, 100, 200, 200], [100, 100, 200, 200]],
    multimask_output=False,
    random_color=False,
)
image.show() # 显示图像
image.save("output.jpg") # 保存图像