segment-anything-model

Segment Anything Model (SAM)

使用 Meta AI 的 Segment Anything Model 进行零样本图像分割的全面指南。

何时使用 SAM

在以下情况下使用 SAM：

• 需要在无需特定任务训练的情况下分割图像中的任意对象
• 构建使用点/框提示词的交互式标注工具
• 为其他视觉模型生成训练数据
• 需要零样本迁移到新的图像领域
• 构建对象检测/分割流水线
• 处理医学、卫星或特定领域的图像

主要特性：

• 零样本分割：无需微调即可适用于任意图像领域
• 灵活的提示词：点、边界框或先前的掩码
• 自动分割：自动生成所有对象掩码
• 高质量：基于 1100 万张图像的 11 亿个掩码训练
• 多种模型尺寸：ViT-B（最快）、ViT-L、ViT-H（最准确）
• ONNX 导出：可部署到浏览器和边缘设备

改用替代方案：

• YOLO/Detectron2：用于带类别的实时对象检测
• Mask2Former：用于带类别的语义/全景分割
• GroundingDINO + SAM：用于文本提示词分割
• SAM 2：用于视频分割任务

快速开始

安装

# From GitHub
pip install git+https://github.com/facebookresearch/segment-anything.git

# Optional dependencies
pip install opencv-python pycocotools matplotlib

# Or use HuggingFace transformers
pip install transformers

下载检查点

# ViT-H (largest, most accurate) - 2.4GB
wget https://dl.fbaipublicfiles.com/segment_anything/sam_vit_h_4b8939.pth

# ViT-L (medium) - 1.2GB
wget https://dl.fbaipublicfiles.com/segment_anything/sam_vit_l_0b3195.pth

# ViT-B (smallest, fastest) - 375MB
wget https://dl.fbaipublicfiles.com/segment_anything/sam_vit_b_01ec64.pth

使用 SamPredictor 的基础用法

import numpy as np
from segment_anything import sam_model_registry, SamPredictor

# Load model
sam = sam_model_registry["vit_h"](checkpoint="sam_vit_h_4b8939.pth")
sam.to(device="cuda")

# Create predictor
predictor = SamPredictor(sam)

# Set image (computes embeddings once)
image = cv2.imread("image.jpg")
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
predictor.set_image(image)

# Predict with point prompts
input_point = np.array([[500, 375]])  # (x, y) coordinates
input_label = np.array([1])  # 1 = foreground, 0 = background

masks, scores, logits = predictor.predict(
    point_coords=input_point,
    point_labels=input_label,
    multimask_output=True  # Returns 3 mask options
)

# Select best mask
best_mask = masks[np.argmax(scores)]

HuggingFace Transformers

import torch
from PIL import Image
from transformers import SamModel, SamProcessor

# Load model and processor
model = SamModel.from_pretrained("facebook/sam-vit-huge")
processor = SamProcessor.from_pretrained("facebook/sam-vit-huge")
model.to("cuda")

# Process image with point prompt
image = Image.open("image.jpg")
input_points = [[[450, 600]]]  # Batch of points

inputs = processor(image, input_points=input_points, return_tensors="pt")
inputs = {k: v.to("cuda") for k, v in inputs.items()}

# Generate masks
with torch.no_grad():
    outputs = model(**inputs)

# Post-process masks to original size
masks = processor.image_processor.post_process_masks(
    outputs.pred_masks.cpu(),
    inputs["original_sizes"].cpu(),
    inputs["reshaped_input_sizes"].cpu()
)

核心概念

模型架构

SAM Architecture:
┌─────────────────┐     ┌─────────────────┐     ┌─────────────────┐
│  Image Encoder  │────▶│ Prompt Encoder  │────▶│  Mask Decoder   │
│     (ViT)       │     │ (Points/Boxes)  │     │ (Transformer)   │
└─────────────────┘     └─────────────────┘     └─────────────────┘
        │                       │                       │
   Image Embeddings      Prompt Embeddings         Masks + IoU
   (computed once)       (per prompt)             predictions

模型变体

模型	检查点	大小	速度	准确率
ViT-H	vit_h	2.4 GB	最慢	最佳
ViT-L	vit_l	1.2 GB	中等	良好
ViT-B	vit_b	375 MB	最快	良好

提示词类型

提示词	描述	使用场景
点（前景）	点击对象	单对象选择
点（背景）	点击对象外部	排除区域
边界框	围绕对象的矩形	较大的对象
先前掩码	低分辨率掩码输入	迭代优化

交互式分割

点提示词

# Single foreground point
input_point = np.array([[500, 375]])
input_label = np.array([1])

masks, scores, logits = predictor.predict(
    point_coords=input_point,
    point_labels=input_label,
    multimask_output=True
)

# Multiple points (foreground + background)
input_points = np.array([[500, 375], [600, 400], [450, 300]])
input_labels = np.array([1, 1, 0])  # 2 foreground, 1 background

masks, scores, logits = predictor.predict(
    point_coords=input_points,
    point_labels=input_labels,
    multimask_output=False  # Single mask when prompts are clear
)

框提示词

# Bounding box [x1, y1, x2, y2]
input_box = np.array([425, 600, 700, 875])

masks, scores, logits = predictor.predict(
    box=input_box,
    multimask_output=False
)

组合提示词

# Box + points for precise control
masks, scores, logits = predictor.predict(
    point_coords=np.array([[500, 375]]),
    point_labels=np.array([1]),
    box=np.array([400, 300, 700, 600]),
    multimask_output=False
)

迭代优化

# Initial prediction
masks, scores, logits = predictor.predict(
    point_coords=np.array([[500, 375]]),
    point_labels=np.array([1]),
    multimask_output=True
)

# Refine with additional point using previous mask
masks, scores, logits = predictor.predict(
    point_coords=np.array([[500, 375], [550, 400]]),
    point_labels=np.array([1, 0]),  # Add background point
    mask_input=logits[np.argmax(scores)][None, :, :],  # Use best mask
    multimask_output=False
)

自动掩码生成

基础自动分割

from segment_anything import SamAutomaticMaskGenerator

# Create generator
mask_generator = SamAutomaticMaskGenerator(sam)

# Generate all masks
masks = mask_generator.generate(image)

# Each mask contains:
# - segmentation: binary mask
# - bbox: [x, y, w, h]
# - area: pixel count
# - predicted_iou: quality score
# - stability_score: robustness score
# - point_coords: generating point

自定义生成

mask_generator = SamAutomaticMaskGenerator(
    model=sam,
    points_per_side=32,          # Grid density (more = more masks)
    pred_iou_thresh=0.88,        # Quality threshold
    stability_score_thresh=0.95,  # Stability threshold
    crop_n_layers=1,             # Multi-scale crops
    crop_n_points_downscale_factor=2,
    min_mask_region_area=100,    # Remove tiny masks
)

masks = mask_generator.generate(image)

过滤掩码

# Sort by area (largest first)
masks = sorted(masks, key=lambda x: x['area'], reverse=True)

# Filter by predicted IoU
high_quality = [m for m in masks if m['predicted_iou'] > 0.9]

# Filter by stability score
stable_masks = [m for m in masks if m['stability_score'] > 0.95]

批处理推理

多张图像

# Process multiple images efficiently
images = [cv2.imread(f"image_{i}.jpg") for i in range(10)]

all_masks = []
for image in images:
    predictor.set_image(image)
    masks, _, _ = predictor.predict(
        point_coords=np.array([[500, 375]]),
        point_labels=np.array([1]),
        multimask_output=True
    )
    all_masks.append(masks)

每张图像多个提示词

# Process multiple prompts efficiently (one image encoding)
predictor.set_image(image)

# Batch of point prompts
points = [
    np.array([[100, 100]]),
    np.array([[200, 200]]),
    np.array([[300, 300]])
]

all_masks = []
for point in points:
    masks, scores, _ = predictor.predict(
        point_coords=point,
        point_labels=np.array([1]),
        multimask_output=True
    )
    all_masks.append(masks[np.argmax(scores)])

ONNX 部署

导出模型

python scripts/export_onnx_model.py \
    --checkpoint sam_vit_h_4b8939.pth \
    --model-type vit_h \
    --output sam_onnx.onnx \
    --return-single-mask

使用 ONNX 模型

import onnxruntime

# Load ONNX model
ort_session = onnxruntime.InferenceSession("sam_onnx.onnx")

# Run inference (image embeddings computed separately)
masks = ort_session.run(
    None,
    {
        "image_embeddings": image_embeddings,
        "point_coords": point_coords,
        "point_labels": point_labels,
        "mask_input": np.zeros((1, 1, 256, 256), dtype=np.float32),
        "has_mask_input": np.array([0], dtype=np.float32),
        "orig_im_size": np.array([h, w], dtype=np.float32)
    }
)

常见工作流

工作流 1：标注工具

import cv2

# Load model
predictor = SamPredictor(sam)
predictor.set_image(image)

def on_click(event, x, y, flags, param):
    if event == cv2.EVENT_LBUTTONDOWN:
        # Foreground point
        masks, scores, _ = predictor.predict(
            point_coords=np.array([[x, y]]),
            point_labels=np.array([1]),
            multimask_output=True
        )
        # Display best mask
        display_mask(masks[np.argmax(scores)])

工作流 2：对象提取

def extract_object(image, point):
    """Extract object at point with transparent background."""
    predictor.set_image(image)

    masks, scores, _ = predictor.predict(
        point_coords=np.array([point]),
        point_labels=np.array([1]),
        multimask_output=True
    )

    best_mask = masks[np.argmax(scores)]

    # Create RGBA output
    rgba = np.zeros((image.shape[0], image.shape[1], 4), dtype=np.uint8)
    rgba[:, :, :3] = image
    rgba[:, :, 3] = best_mask * 255

    return rgba

工作流 3：医学图像分割

# Process medical images (grayscale to RGB)
medical_image = cv2.imread("scan.png", cv2.IMREAD_GRAYSCALE)
rgb_image = cv2.cvtColor(medical_image, cv2.COLOR_GRAY2RGB)

predictor.set_image(rgb_image)

# Segment region of interest
masks, scores, _ = predictor.predict(
    box=np.array([x1, y1, x2, y2]),  # ROI bounding box
    multimask_output=True
)

输出格式

掩码数据结构

# SamAutomaticMaskGenerator output
{
    "segmentation": np.ndarray,  # H×W binary mask
    "bbox": [x, y, w, h],        # Bounding box
    "area": int,                 # Pixel count
    "predicted_iou": float,      # 0-1 quality score
    "stability_score": float,    # 0-1 robustness score
    "crop_box": [x, y, w, h],    # Generation crop region
    "point_coords": [[x, y]],    # Input point
}

COCO RLE 格式

from pycocotools import mask as mask_utils

# Encode mask to RLE
rle = mask_utils.encode(np.asfortranarray(mask.astype(np.uint8)))
rle["counts"] = rle["counts"].decode("utf-8")

# Decode RLE to mask
decoded_mask = mask_utils.decode(rle)

性能优化

GPU 内存

# Use smaller model for limited VRAM
sam = sam_model_registry["vit_b"](checkpoint="sam_vit_b_01ec64.pth")

# Process images in batches
# Clear CUDA cache between large batches
torch.cuda.empty_cache()

速度优化

# Use half precision
sam = sam.half()

# Reduce points for automatic generation
mask_generator = SamAutomaticMaskGenerator(
    model=sam,
    points_per_side=16,  # Default is 32
)

# Use ONNX for deployment
# Export with --return-single-mask for faster inference

常见问题

问题	解决方案
内存不足	使用 ViT-B 模型，缩小图像尺寸
推理缓慢	使用 ViT-B，减少 points_per_side
掩码质量差	尝试不同的提示词，使用框 + 点
边缘伪影	使用 stability_score 过滤
遗漏小对象	增加 points_per_side

参考资料

• 高级用法 - 批处理、微调、集成
• 故障排除 - 常见问题与解决方案

资源

• GitHub：https://github.com/facebookresearch/segment-anything
• 论文：https://arxiv.org/abs/2304.02643
• 演示：https://segment-anything.com
• SAM 2（视频）：https://github.com/facebookresearch/segment-anything-2
• HuggingFace：https://huggingface.co/facebook/sam-vit-huge