Segment Anything model部署和使用

? ? ? ?分割一切模型 （SAM） 根据输入提示（如点或框）生成高质量的对象mask，并可用于为图像中的所有对象生成mask。它已经在 1100 万张图像和 11 亿个掩码的数据集上进行了训练，在各种分割任务上具有强大的零样本性能。官方文档：facebookresearch/segment-anything: The repository provides code for running inference with the SegmentAnything Model (SAM), links for downloading the trained model checkpoints, and example notebooks that show how to use the model. (github.com)

Installation

代码能够运行需要：python>=3.8, as well as?pytorch>=1.7?and?torchvision>=0.8

1.（可以先创建一个虚拟环境如sam）下载sam：

（建议直接下载，解压到当前文件夹，并把解压出的文件夹名字改成segment-anything）

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

?

2.安装：

git clone git@github.com:facebookresearch/segment-anything.git
cd segment-anything; pip install -e .

下载zip文件的解压后

cd segment-anything#激活虚拟环境进入文件夹

pip install -e .

??如何安装其他依赖：torch等

pip install opencv-python pycocotools matplotlib onnxruntime onnx

?3.下载权重模型

比如vit-h

https://dl.fbaipublicfiles.com/segment_anything/sam_vit_h_4b8939.pth

建议直接下载进入segment-anything文件里面

各个模块介绍

?

?Sam

1.图像编码器（image_encoder）：使用ViT模型对图像进行编码，提取图像的特征。

2.提示编码器（prompt_encoder）：将图像中的区域位置编码成向量，并与提示文本进行拼接，形成提示编码器的输入。

3.掩膜解码器（mask_decoder）：将提示编码器的输出作为输入，生成掩膜，用于对图像进行分割。

在函数的参数中，encoder_embed_dim、encoder_depth、encoder_num_heads、encoder_global_attn_indexes等是图像编码器使用的参数，用于控制ViT模型的参数；

# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.

# This source code is licensed under the license found in the
# LICENSE file in the root directory of this source tree.

import torch
from torch import nn
from torch.nn import functional as F

from typing import Any, Dict, List, Tuple

from .image_encoder import ImageEncoderViT
from .mask_decoder import MaskDecoder
from .prompt_encoder import PromptEncoder


class Sam(nn.Module):
    mask_threshold: float = 0.0
    image_format: str = "RGB"

    def __init__(
        self,
        image_encoder: ImageEncoderViT,
        prompt_encoder: PromptEncoder,
        mask_decoder: MaskDecoder,
        pixel_mean: List[float] = [123.675, 116.28, 103.53],
        pixel_std: List[float] = [58.395, 57.12, 57.375],
    ) -> None:
        """
        SAM predicts object masks from an image and input prompts.

        Arguments:
          image_encoder (ImageEncoderViT): The backbone used to encode the
            image into image embeddings that allow for efficient mask prediction.
          prompt_encoder (PromptEncoder): Encodes various types of input prompts.
          mask_decoder (MaskDecoder): Predicts masks from the image embeddings
            and encoded prompts.
          pixel_mean (list(float)): Mean values for normalizing pixels in the input image.
          pixel_std (list(float)): Std values for normalizing pixels in the input image.
        """
        super().__init__()
        self.image_encoder = image_encoder
        self.prompt_encoder = prompt_encoder
        self.mask_decoder = mask_decoder
        self.register_buffer("pixel_mean", torch.Tensor(pixel_mean).view(-1, 1, 1), False)
        self.register_buffer("pixel_std", torch.Tensor(pixel_std).view(-1, 1, 1), False)

    @property
    def device(self) -> Any:
        return self.pixel_mean.device
    
# Sam的forword要求输入的是list对象，其对图像编码时是进行batch操作（一次性推理出所有图像的特征）
# 而对提示输入和mask生成则是单独操作（通过for循环预测每一个位置提示所对应的mask）；其输出也是list对象
# 每个元素包含masks、iou_predictions和low_res_logits值
# 其中，masks是low_res_logits的高分辨率结果，并按照mask_threshold进行二值化（也就是说模型只时预测出低分辨率的low_res_logits）
    @torch.no_grad()
    def forward(
        self,
        batched_input: List[Dict[str, Any]],
        multimask_output: bool,
    ) -> List[Dict[str, torch.Tensor]]:
        """
        Predicts masks end-to-end from provided images and prompts.
        If prompts are not known in advance, using SamPredictor is
        recommended over calling the model directly.

        Arguments:
          batched_input (list(dict)): A list over input images, each a
            dictionary with the following keys. A prompt key can be
            excluded if it is not present.
              'image': The image as a torch tensor in 3xHxW format,
                already transformed for input to the model.
              'original_size': (tuple(int, int)) The original size of
                the image before transformation, as (H, W).
              'point_coords': (torch.Tensor) Batched point prompts for
                this image, with shape BxNx2. Already transformed to the
                input frame of the model.
              'point_labels': (torch.Tensor) Batched labels for point prompts,
                with shape BxN.
              'boxes': (torch.Tensor) Batched box inputs, with shape Bx4.
                Already transformed to the input frame of the model.
              'mask_inputs': (torch.Tensor) Batched mask inputs to the model,
                in the form Bx1xHxW.
          multimask_output (bool): Whether the model should predict multiple
            disambiguating masks, or return a single mask.

        Returns:
          (list(dict)): A list over input images, where each element is
            as dictionary with the following keys.
              'masks': (torch.Tensor) Batched binary mask predictions,
                with shape BxCxHxW, where B is the number of input prompts,
                C is determined by multimask_output, and (H, W) is the
                original size of the image.
              'iou_predictions': (torch.Tensor) The model's predictions
                of mask quality, in shape BxC.
              'low_res_logits': (torch.Tensor) Low resolution logits with
                shape BxCxHxW, where H=W=256. Can be passed as mask input
                to subsequent iterations of prediction.
        """
        input_images = torch.stack([self.preprocess(x["image"]) for x in batched_input], dim=0)
        image_embeddings = self.image_encoder(input_images)

        outputs = []
        for image_record, curr_embedding in zip(batched_input, image_embeddings):
            if "point_coords" in image_record:
                points = (image_record["point_coords"], image_record["point_labels"])
            else:
                points = None
            sparse_embeddings, dense_embeddings = self.prompt_encoder(
                points=points,
                boxes=image_record.get("boxes", None),
                masks=image_record.get("mask_inputs", None),
            )
            low_res_masks, iou_predictions = self.mask_decoder(
                image_embeddings=curr_embedding.unsqueeze(0),
                image_pe=self.prompt_encoder.get_dense_pe(),
                sparse_prompt_embeddings=sparse_embeddings,
                dense_prompt_embeddings=dense_embeddings,
                multimask_output=multimask_output,
            )
            masks = self.postprocess_masks(
                low_res_masks,
                input_size=image_record["image"].shape[-2:],
                original_size=image_record["original_size"],
            )
            masks = masks > self.mask_threshold
            outputs.append(
                {
                    "masks": masks,
                    "iou_predictions": iou_predictions,
                    "low_res_logits": low_res_masks,
                }
            )
        return outputs

    def postprocess_masks(
        self,
        masks: torch.Tensor,
        input_size: Tuple[int, ...],
        original_size: Tuple[int, ...],
    ) -> torch.Tensor:
        """
        Remove padding and upscale masks to the original image size.

        Arguments:
          masks (torch.Tensor): Batched masks from the mask_decoder,
            in BxCxHxW format.
          input_size (tuple(int, int)): The size of the image input to the
            model, in (H, W) format. Used to remove padding.
          original_size (tuple(int, int)): The original size of the image
            before resizing for input to the model, in (H, W) format.

        Returns:
          (torch.Tensor): Batched masks in BxCxHxW format, where (H, W)
            is given by original_size.
        """
        masks = F.interpolate(
            masks,
            (self.image_encoder.img_size, self.image_encoder.img_size),
            mode="bilinear",
            align_corners=False,
        )
        masks = masks[..., : input_size[0], : input_size[1]]
        masks = F.interpolate(masks, original_size, mode="bilinear", align_corners=False)
        return masks

    def preprocess(self, x: torch.Tensor) -> torch.Tensor:
        """Normalize pixel values and pad to a square input."""
        # Normalize colors
        x = (x - self.pixel_mean) / self.pixel_std

        # Pad
        h, w = x.shape[-2:]
        padh = self.image_encoder.img_size - h
        padw = self.image_encoder.img_size - w
        x = F.pad(x, (0, padw, 0, padh))
        return x

image_encoder模块?

利用mae预训练的vit，最低限度适应高分辨率的输入，该encoder在prompt encoder之前，对每张图像只运行一次。??输入(c,h,w)的图像，对图像进行缩放，按照长边缩放成1024，短边不够就pad,得(c,1024,1024)的图像，经过image encoder，得到对图像16倍下采样的feature，大小为(256,64,64)。

# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.

# This source code is licensed under the license found in the
# LICENSE file in the root directory of this source tree.

import torch
import torch.nn as nn
import torch.nn.functional as F

from typing import Optional, Tuple, Type

from .common import LayerNorm2d, MLPBlock

# 其先由patch_embed对输入数据进行16倍的下采样（将patch_size设为16）
# 并将embed_dim[token长度]设为768（原始的vit中embed_dim是为196=img_wimg_h/(patch_wpatch_h)，img_w:224 patch_w:16）
# 这与原始ViT相比，是存在一定信息缺失的（64x64=4096）
#ImageEncoderViT还多了一个neck层，用于将embed_dim从768转换到所需的out_chans（256）
# This class and its supporting functions below lightly adapted from the ViTDet backbone available at: https://github.com/facebookresearch/detectron2/blob/main/detectron2/modeling/backbone/vit.py # noqa
class ImageEncoderViT(nn.Module):
    def __init__(
        self,
        img_size: int = 1024,
        patch_size: int = 16,
        in_chans: int = 3,
        embed_dim: int = 768,
        depth: int = 12,
        num_heads: int = 12,
        mlp_ratio: float = 4.0,
        out_chans: int = 256,
        qkv_bias: bool = True,
        norm_layer: Type[nn.Module] = nn.LayerNorm,
        act_layer: Type[nn.Module] = nn.GELU,
        use_abs_pos: bool = True,
        use_rel_pos: bool = False,
        rel_pos_zero_init: bool = True,
        window_size: int = 0,
        global_attn_indexes: Tuple[int, ...] = (),
    ) -> None:
        """
        Args:
            img_size (int): Input image size.
            patch_size (int): Patch size.
            in_chans (int): Number of input image channels.
            embed_dim (int): Patch embedding dimension.
            depth (int): Depth of ViT.
            num_heads (int): Number of attention heads in each ViT block.
            mlp_ratio (float): Ratio of mlp hidden dim to embedding dim.
            qkv_bias (bool): If True, add a learnable bias to query, key, value.
            norm_layer (nn.Module): Normalization layer.
            act_layer (nn.Module): Activation layer.
            use_abs_pos (bool): If True, use absolute positional embeddings.
            use_rel_pos (bool): If True, add relative positional embeddings to the attention map.
            rel_pos_zero_init (bool): If True, zero initialize relative positional parameters.
            window_size (int): Window size for window attention blocks.
            global_attn_indexes (list): Indexes for blocks using global attention.
        """
        """
      img_size：输入图像的大小（宽度和高度）。
      patch_size：图像分块的大小。
      in_chans：输入图像的通道数。
      embed_dim：每个图像块的向量表示的维度。
      depth：ViT 模型的深度，即 ViT 中包含的 Transformer block 的数量。
      num_heads：每个 Transformer block 中多头自注意力机制中的注意力头数。
      mlp_ratio：MLP 线性变换隐藏层维度与向量表示维度之比。
      qkv_bias：如果为 True，则给查询、键、值添加可学习偏置。
      norm_layer：规范化层。
      act_layer：激活函数。
      use_abs_pos：如果为 True，则使用绝对位置嵌入。
      use_rel_pos：如果为 True，则在注意力矩阵中加入相对位置嵌入。
      rel_pos_zero_init：如果为 True，则将相对位置参数初始化为零。
      window_size：窗口注意力块中窗口的大小。
      global_attn_indexes：使用全局注意力的 Transformer block 的索引列表。
        """
        super().__init__()
        self.img_size = img_size

        self.patch_embed = PatchEmbed(
            kernel_size=(patch_size, patch_size),
            stride=(patch_size, patch_size),
            in_chans=in_chans,
            embed_dim=embed_dim,
        )

        self.pos_embed: Optional[nn.Parameter] = None
        if use_abs_pos:
            # Initialize absolute positional embedding with pretrain image size.
            self.pos_embed = nn.Parameter(
                torch.zeros(1, img_size // patch_size, img_size // patch_size, embed_dim)
            )

        self.blocks = nn.ModuleList()
        for i in range(depth):
            block = Block(
                dim=embed_dim,
                num_heads=num_heads,
                mlp_ratio=mlp_ratio,
                qkv_bias=qkv_bias,
                norm_layer=norm_layer,
                act_layer=act_layer,
                use_rel_pos=use_rel_pos,
                rel_pos_zero_init=rel_pos_zero_init,
                window_size=window_size if i not in global_attn_indexes else 0,
                input_size=(img_size // patch_size, img_size // patch_size),
            )
            self.blocks.append(block)

        self.neck = nn.Sequential(
            nn.Conv2d(
                embed_dim,
                out_chans,
                kernel_size=1,
                bias=False,
            ),
            LayerNorm2d(out_chans),
            nn.Conv2d(
                out_chans,
                out_chans,
                kernel_size=3,
                padding=1,
                bias=False,
            ),
            LayerNorm2d(out_chans),
        )

    def forward(self, x: torch.Tensor) -> torch.Tensor:
        x = self.patch_embed(x)
        if self.pos_embed is not None:
            x = x + self.pos_embed

        for blk in self.blocks:
            x = blk(x)

        x = self.neck(x.permute(0, 3, 1, 2))

        return x


class Block(nn.Module):
    """Transformer blocks with support of window attention and residual propagation blocks"""

    def __init__(
        self,
        dim: int,
        num_heads: int,
        mlp_ratio: float = 4.0,
        qkv_bias: bool = True,
        norm_layer: Type[nn.Module] = nn.LayerNorm,
        act_layer: Type[nn.Module] = nn.GELU,
        use_rel_pos: bool = False,
        rel_pos_zero_init: bool = True,
        window_size: int = 0,
        input_size: Optional[Tuple[int, int]] = None,
    ) -> None:
        """
        Args:
            dim (int): Number of input channels.
            num_heads (int): Number of attention heads in each ViT block.
            mlp_ratio (float): Ratio of mlp hidden dim to embedding dim.
            qkv_bias (bool): If True, add a learnable bias to query, key, value.
            norm_layer (nn.Module): Normalization layer.
            act_layer (nn.Module): Activation layer.
            use_rel_pos (bool): If True, add relative positional embeddings to the attention map.
            rel_pos_zero_init (bool): If True, zero initialize relative positional parameters.
            window_size (int): Window size for window attention blocks. If it equals 0, then
                use global attention.
            input_size (tuple(int, int) or None): Input resolution for calculating the relative
                positional parameter size.
        """
        super().__init__()
        self.norm1 = norm_layer(dim)
        self.attn = Attention(
            dim,
            num_heads=num_heads,
            qkv_bias=qkv_bias,
            use_rel_pos=use_rel_pos,
            rel_pos_zero_init=rel_pos_zero_init,
            input_size=input_size if window_size == 0 else (window_size, window_size),
        )

        self.norm2 = norm_layer(dim)
        self.mlp = MLPBlock(embedding_dim=dim, mlp_dim=int(dim * mlp_ratio), act=act_layer)

        self.window_size = window_size

    def forward(self, x: torch.Tensor) -> torch.Tensor:
        shortcut = x
        x = self.norm1(x)
        # Window partition
        if self.window_size > 0:
            H, W = x.shape[1], x.shape[2]
            x, pad_hw = window_partition(x, self.window_size)

        x = self.attn(x)
        # Reverse window partition
        if self.window_size > 0:
            x = window_unpartition(x, self.window_size, pad_hw, (H, W))

        x = shortcut + x
        x = x + self.mlp(self.norm2(x))

        return x


class Attention(nn.Module):
    """Multi-head Attention block with relative position embeddings."""

    def __init__(
        self,
        dim: int,
        num_heads: int = 8,
        qkv_bias: bool = True,
        use_rel_pos: bool = False,
        rel_pos_zero_init: bool = True,
        input_size: Optional[Tuple[int, int]] = None,
    ) -> None:
        """
        Args:
            dim (int): Number of input channels.
            num_heads (int): Number of attention heads.
            qkv_bias (bool):  If True, add a learnable bias to query, key, value.
            rel_pos (bool): If True, add relative positional embeddings to the attention map.
            rel_pos_zero_init (bool): If True, zero initialize relative positional parameters.
            input_size (tuple(int, int) or None): Input resolution for calculating the relative
                positional parameter size.
        """
        super().__init__()
        self.num_heads = num_heads
        head_dim = dim // num_heads
        self.scale = head_dim**-0.5

        self.qkv = nn.Linear(dim, dim * 3, bias=qkv_bias)
        self.proj = nn.Linear(dim, dim)

        self.use_rel_pos = use_rel_pos
        if self.use_rel_pos:
            assert (
                input_size is not None
            ), "Input size must be provided if using relative positional encoding."
            # initialize relative positional embeddings
            self.rel_pos_h = nn.Parameter(torch.zeros(2 * input_size[0] - 1, head_dim))
            self.rel_pos_w = nn.Parameter(torch.zeros(2 * input_size[1] - 1, head_dim))

    def forward(self, x: torch.Tensor) -> torch.Tensor:
        B, H, W, _ = x.shape
        # qkv with shape (3, B, nHead, H * W, C)
        qkv = self.qkv(x).reshape(B, H * W, 3, self.num_heads, -1).permute(2, 0, 3, 1, 4)
        # q, k, v with shape (B * nHead, H * W, C)
        q, k, v = qkv.reshape(3, B * self.num_heads, H * W, -1).unbind(0)

        attn = (q * self.scale) @ k.transpose(-2, -1)

        if self.use_rel_pos:
            attn = add_decomposed_rel_pos(attn, q, self.rel_pos_h, self.rel_pos_w, (H, W), (H, W))

        attn = attn.softmax(dim=-1)
        x = (attn @ v).view(B, self.num_heads, H, W, -1).permute(0, 2, 3, 1, 4).reshape(B, H, W, -1)
        x = self.proj(x)

        return x


def window_partition(x: torch.Tensor, window_size: int) -> Tuple[torch.Tensor, Tuple[int, int]]:
    """
    Partition into non-overlapping windows with padding if needed.
    Args:
        x (tensor): input tokens with [B, H, W, C].
        window_size (int): window size.

    Returns:
        windows: windows after partition with [B * num_windows, window_size, window_size, C].
        (Hp, Wp): padded height and width before partition
    """
    B, H, W, C = x.shape

    pad_h = (window_size - H % window_size) % window_size
    pad_w = (window_size - W % window_size) % window_size
    if pad_h > 0 or pad_w > 0:
        x = F.pad(x, (0, 0, 0, pad_w, 0, pad_h))
    Hp, Wp = H + pad_h, W + pad_w

    x = x.view(B, Hp // window_size, window_size, Wp // window_size, window_size, C)
    windows = x.permute(0, 1, 3, 2, 4, 5).contiguous().view(-1, window_size, window_size, C)
    return windows, (Hp, Wp)


def window_unpartition(
    windows: torch.Tensor, window_size: int, pad_hw: Tuple[int, int], hw: Tuple[int, int]
) -> torch.Tensor:
    """
    Window unpartition into original sequences and removing padding.
    Args:
        windows (tensor): input tokens with [B * num_windows, window_size, window_size, C].
        window_size (int): window size.
        pad_hw (Tuple): padded height and width (Hp, Wp).
        hw (Tuple): original height and width (H, W) before padding.

    Returns:
        x: unpartitioned sequences with [B, H, W, C].
    """
    Hp, Wp = pad_hw
    H, W = hw
    B = windows.shape[0] // (Hp * Wp // window_size // window_size)
    x = windows.view(B, Hp // window_size, Wp // window_size, window_size, window_size, -1)
    x = x.permute(0, 1, 3, 2, 4, 5).contiguous().view(B, Hp, Wp, -1)

    if Hp > H or Wp > W:
        x = x[:, :H, :W, :].contiguous()
    return x


def get_rel_pos(q_size: int, k_size: int, rel_pos: torch.Tensor) -> torch.Tensor:
    """
    Get relative positional embeddings according to the relative positions of
        query and key sizes.
    Args:
        q_size (int): size of query q.
        k_size (int): size of key k.
        rel_pos (Tensor): relative position embeddings (L, C).

    Returns:
        Extracted positional embeddings according to relative positions.
    """
    max_rel_dist = int(2 * max(q_size, k_size) - 1)
    # Interpolate rel pos if needed.
    if rel_pos.shape[0] != max_rel_dist:
        # Interpolate rel pos.
        rel_pos_resized = F.interpolate(
            rel_pos.reshape(1, rel_pos.shape[0], -1).permute(0, 2, 1),
            size=max_rel_dist,
            mode="linear",
        )
        rel_pos_resized = rel_pos_resized.reshape(-1, max_rel_dist).permute(1, 0)
    else:
        rel_pos_resized = rel_pos

    # Scale the coords with short length if shapes for q and k are different.
    q_coords = torch.arange(q_size)[:, None] * max(k_size / q_size, 1.0)
    k_coords = torch.arange(k_size)[None, :] * max(q_size / k_size, 1.0)
    relative_coords = (q_coords - k_coords) + (k_size - 1) * max(q_size / k_size, 1.0)

    return rel_pos_resized[relative_coords.long()]


def add_decomposed_rel_pos(
    attn: torch.Tensor,
    q: torch.Tensor,
    rel_pos_h: torch.Tensor,
    rel_pos_w: torch.Tensor,
    q_size: Tuple[int, int],
    k_size: Tuple[int, int],
) -> torch.Tensor:
    """
    Calculate decomposed Relative Positional Embeddings from :paper:`mvitv2`.
    https://github.com/facebookresearch/mvit/blob/19786631e330df9f3622e5402b4a419a263a2c80/mvit/models/attention.py   # noqa B950
    Args:
        attn (Tensor): attention map.
        q (Tensor): query q in the attention layer with shape (B, q_h * q_w, C).
        rel_pos_h (Tensor): relative position embeddings (Lh, C) for height axis.
        rel_pos_w (Tensor): relative position embeddings (Lw, C) for width axis.
        q_size (Tuple): spatial sequence size of query q with (q_h, q_w).
        k_size (Tuple): spatial sequence size of key k with (k_h, k_w).

    Returns:
        attn (Tensor): attention map with added relative positional embeddings.
    """
    q_h, q_w = q_size
    k_h, k_w = k_size
    Rh = get_rel_pos(q_h, k_h, rel_pos_h)
    Rw = get_rel_pos(q_w, k_w, rel_pos_w)

    B, _, dim = q.shape
    r_q = q.reshape(B, q_h, q_w, dim)
    rel_h = torch.einsum("bhwc,hkc->bhwk", r_q, Rh)
    rel_w = torch.einsum("bhwc,wkc->bhwk", r_q, Rw)

    attn = (
        attn.view(B, q_h, q_w, k_h, k_w) + rel_h[:, :, :, :, None] + rel_w[:, :, :, None, :]
    ).view(B, q_h * q_w, k_h * k_w)

    return attn


class PatchEmbed(nn.Module):
    """
    Image to Patch Embedding.
    """

    def __init__(
        self,
        kernel_size: Tuple[int, int] = (16, 16),
        stride: Tuple[int, int] = (16, 16),
        padding: Tuple[int, int] = (0, 0),
        in_chans: int = 3,
        embed_dim: int = 768,
    ) -> None:
        """
        Args:
            kernel_size (Tuple): kernel size of the projection layer.
            stride (Tuple): stride of the projection layer.
            padding (Tuple): padding size of the projection layer.
            in_chans (int): Number of input image channels.
            embed_dim (int): Patch embedding dimension.
        """
        super().__init__()

        self.proj = nn.Conv2d(
            in_chans, embed_dim, kernel_size=kernel_size, stride=stride, padding=padding
        )

    def forward(self, x: torch.Tensor) -> torch.Tensor:
        x = self.proj(x)
        # B C H W -> B H W C
        x = x.permute(0, 2, 3, 1)
        return x

prompt_encoder模块?

PromptEncoder用于对输入模型的points、boxes和masks信息进行编码，将其统一为空间特征编码的格式。其主体代码如下所示,可见其编码器并不复杂，属于轻量化的结构。其对points、boxes和masks编码时允许有部分值空缺（空缺使用默认值），其将points和boxes组装为sparse_embeddings，将mask组装为dense_embeddings其对mask的采样由多个attention层实现。

# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.

# This source code is licensed under the license found in the
# LICENSE file in the root directory of this source tree.

import numpy as np
import torch
from torch import nn

from typing import Any, Optional, Tuple, Type

from .common import LayerNorm2d


class PromptEncoder(nn.Module):
    def __init__(
        self,
        embed_dim: int,
        image_embedding_size: Tuple[int, int],
        input_image_size: Tuple[int, int],
        mask_in_chans: int,
        activation: Type[nn.Module] = nn.GELU,
    ) -> None:
        """
        Encodes prompts for input to SAM's mask decoder.

        Arguments:
          embed_dim (int): The prompts' embedding dimension
          image_embedding_size (tuple(int, int)): The spatial size of the
            image embedding, as (H, W).
          input_image_size (int): The padded size of the image as input
            to the image encoder, as (H, W).
          mask_in_chans (int): The number of hidden channels used for
            encoding input masks.
          activation (nn.Module): The activation to use when encoding
            input masks.
        """
        super().__init__()
        self.embed_dim = embed_dim
        self.input_image_size = input_image_size
        self.image_embedding_size = image_embedding_size
        self.pe_layer = PositionEmbeddingRandom(embed_dim // 2)

        self.num_point_embeddings: int = 4  # pos/neg point + 2 box corners
        point_embeddings = [nn.Embedding(1, embed_dim) for i in range(self.num_point_embeddings)]
        self.point_embeddings = nn.ModuleList(point_embeddings)
        self.not_a_point_embed = nn.Embedding(1, embed_dim)

        self.mask_input_size = (4 * image_embedding_size[0], 4 * image_embedding_size[1])
        self.mask_downscaling = nn.Sequential(
            nn.Conv2d(1, mask_in_chans // 4, kernel_size=2, stride=2),
            LayerNorm2d(mask_in_chans // 4),
            activation(),
            nn.Conv2d(mask_in_chans // 4, mask_in_chans, kernel_size=2, stride=2),
            LayerNorm2d(mask_in_chans),
            activation(),
            nn.Conv2d(mask_in_chans, embed_dim, kernel_size=1),
        )
        self.no_mask_embed = nn.Embedding(1, embed_dim)

    def get_dense_pe(self) -> torch.Tensor:
        """
        Returns the positional encoding used to encode point prompts,
        applied to a dense set of points the shape of the image encoding.

        Returns:
          torch.Tensor: Positional encoding with shape
            1x(embed_dim)x(embedding_h)x(embedding_w)
        """
        return self.pe_layer(self.image_embedding_size).unsqueeze(0)
#PromptEncoder对于points附加的的labels信息操作为：-1 表示ignore，0 表示 负点（对应pe_layer输出的【0】），1 表示正点（对应pe_layer输出的【1】）
    def _embed_points(
        self,
        points: torch.Tensor,
        labels: torch.Tensor,
        pad: bool,
    ) -> torch.Tensor:
        """Embeds point prompts."""
        points = points + 0.5  # Shift to center of pixel
        if pad:
            padding_point = torch.zeros((points.shape[0], 1, 2), device=points.device)
            padding_label = -torch.ones((labels.shape[0], 1), device=labels.device)
            points = torch.cat([points, padding_point], dim=1)
            labels = torch.cat([labels, padding_label], dim=1)
        point_embedding = self.pe_layer.forward_with_coords(points, self.input_image_size)
        point_embedding[labels == -1] = 0.0
        point_embedding[labels == -1] += self.not_a_point_embed.weight
        point_embedding[labels == 0] += self.point_embeddings[0].weight
        point_embedding[labels == 1] += self.point_embeddings[1].weight
        return point_embedding
#PromptEncoder对于points和boxes都使用PositionEmbeddingRandom（pe_layer）进行编码，可以看到其将boxes转换为2个点然后输入了模型。point输入，则对应着pe_layer输出的【2和3】
    def _embed_boxes(self, boxes: torch.Tensor) -> torch.Tensor:
        """Embeds box prompts."""
        boxes = boxes + 0.5  # Shift to center of pixel
        coords = boxes.reshape(-1, 2, 2)
        corner_embedding = self.pe_layer.forward_with_coords(coords, self.input_image_size)
        corner_embedding[:, 0, :] += self.point_embeddings[2].weight
        corner_embedding[:, 1, :] += self.point_embeddings[3].weight
        return corner_embedding

    def _embed_masks(self, masks: torch.Tensor) -> torch.Tensor:
        """Embeds mask inputs."""
        mask_embedding = self.mask_downscaling(masks)
        return mask_embedding

    def _get_batch_size(
        self,
        points: Optional[Tuple[torch.Tensor, torch.Tensor]],
        boxes: Optional[torch.Tensor],
        masks: Optional[torch.Tensor],
    ) -> int:
        """
        Gets the batch size of the output given the batch size of the input prompts.
        """
        if points is not None:
            return points[0].shape[0]
        elif boxes is not None:
            return boxes.shape[0]
        elif masks is not None:
            return masks.shape[0]
        else:
            return 1

    def _get_device(self) -> torch.device:
        return self.point_embeddings[0].weight.device

    def forward(
        self,
        points: Optional[Tuple[torch.Tensor, torch.Tensor]],
        boxes: Optional[torch.Tensor],
        masks: Optional[torch.Tensor],
    ) -> Tuple[torch.Tensor, torch.Tensor]:
        """
        Embeds different types of prompts, returning both sparse and dense
        embeddings.

        Arguments:
          points (tuple(torch.Tensor, torch.Tensor) or none): point coordinates
            and labels to embed.
          boxes (torch.Tensor or none): boxes to embed
          masks (torch.Tensor or none): masks to embed

        Returns:
          torch.Tensor: sparse embeddings for the points and boxes, with shape
            BxNx(embed_dim), where N is determined by the number of input points
            and boxes.
          torch.Tensor: dense embeddings for the masks, in the shape
            Bx(embed_dim)x(embed_H)x(embed_W)
        """
        bs = self._get_batch_size(points, boxes, masks)
        sparse_embeddings = torch.empty((bs, 0, self.embed_dim), device=self._get_device())
        if points is not None:
            coords, labels = points
            point_embeddings = self._embed_points(coords, labels, pad=(boxes is None))
            sparse_embeddings = torch.cat([sparse_embeddings, point_embeddings], dim=1)
        if boxes is not None:
            box_embeddings = self._embed_boxes(boxes)
            sparse_embeddings = torch.cat([sparse_embeddings, box_embeddings], dim=1)

        if masks is not None:
            dense_embeddings = self._embed_masks(masks)
        else:
            dense_embeddings = self.no_mask_embed.weight.reshape(1, -1, 1, 1).expand(
                bs, -1, self.image_embedding_size[0], self.image_embedding_size[1]
            )

        return sparse_embeddings, dense_embeddings

#PositionEmbeddingRandom 的实现如下，其核心是_pe_encoding函数，对坐标进行标准化，然后乘上一个可训练参数，再去sin和cos做编码
    
class PositionEmbeddingRandom(nn.Module):
    """
    Positional encoding using random spatial frequencies.
    """

    def __init__(self, num_pos_feats: int = 64, scale: Optional[float] = None) -> None:
        super().__init__()
        if scale is None or scale <= 0.0:
            scale = 1.0
        self.register_buffer(
            "positional_encoding_gaussian_matrix",
            scale * torch.randn((2, num_pos_feats)),
        )

    def _pe_encoding(self, coords: torch.Tensor) -> torch.Tensor:
        """Positionally encode points that are normalized to [0,1]."""
        # assuming coords are in [0, 1]^2 square and have d_1 x ... x d_n x 2 shape
        coords = 2 * coords - 1
        coords = coords @ self.positional_encoding_gaussian_matrix
        coords = 2 * np.pi * coords
        # outputs d_1 x ... x d_n x C shape
        return torch.cat([torch.sin(coords), torch.cos(coords)], dim=-1)

    def forward(self, size: Tuple[int, int]) -> torch.Tensor:
        """Generate positional encoding for a grid of the specified size."""
        h, w = size
        device: Any = self.positional_encoding_gaussian_matrix.device
        grid = torch.ones((h, w), device=device, dtype=torch.float32)
        y_embed = grid.cumsum(dim=0) - 0.5
        x_embed = grid.cumsum(dim=1) - 0.5
        y_embed = y_embed / h
        x_embed = x_embed / w

        pe = self._pe_encoding(torch.stack([x_embed, y_embed], dim=-1))
        return pe.permute(2, 0, 1)  # C x H x W

    def forward_with_coords(
        self, coords_input: torch.Tensor, image_size: Tuple[int, int]
    ) -> torch.Tensor:
        """Positionally encode points that are not normalized to [0,1]."""
        coords = coords_input.clone()
        coords[:, :, 0] = coords[:, :, 0] / image_size[1]
        coords[:, :, 1] = coords[:, :, 1] / image_size[0]
        return self._pe_encoding(coords.to(torch.float))  # B x N x C

Mask_Decoder模块?

在prompt embeddings中插入一个可学习的token，用于docoder的输出。

（1）prompt toekns+output tokens进行self attn,

（2）用得到的token和image embedding进行 cross attn（token作为Q）

（3）point-wise MLP 更新token

（4）用image embedding和（3）的token进行cross atten（image embedding作为Q）

重复上述步骤2次，再将attn再通过残差进行连接，最终输出masks和iou scores。

为了解决输出模糊性问题(一个提示可能生成多个mask，比如衣服上的一个点，既可以表示衣服，也表示穿衣服的人)，预测输出多个masks(发现**整体，部分，子部分**已经足够描述mask)，在训练过程中,只回传最小的loss，为了对mask进行排序，增加一个小的head预测mask和目标的iou。

当输入多个提示时，生成的mask会比较接近，为了减少loss退化和确保获取明确的mask，此时只预测一个mask（作为第4个预测mask，只有多个提示时才预测，当单个提示时不用）

# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.

# This source code is licensed under the license found in the
# LICENSE file in the root directory of this source tree.

import torch
from torch import nn
from torch.nn import functional as F

from typing import List, Tuple, Type

from .common import LayerNorm2d
# 用于根据PromptEncoder和ImageEncoderViT的输入生成mask，其是一种transformer架构的解码头
#有两个输出头：output_upscaling和iou_prediction_head，两个头输出的数量是一样的。
# 其forward函数仅是对输出结果进行了选择操作，核心推理是由predict_masks函数和transformer对象


class MaskDecoder(nn.Module):
    def __init__(
        self,
        *,
        transformer_dim: int,
        transformer: nn.Module,
        num_multimask_outputs: int = 3,
        activation: Type[nn.Module] = nn.GELU,
        iou_head_depth: int = 3,
        iou_head_hidden_dim: int = 256,
    ) -> None:
        """
        Predicts masks given an image and prompt embeddings, using a
        transformer architecture.

        Arguments:
          transformer_dim (int): the channel dimension of the transformer
          transformer (nn.Module): the transformer used to predict masks
          num_multimask_outputs (int): the number of masks to predict
            when disambiguating masks
          activation (nn.Module): the type of activation to use when
            upscaling masks
          iou_head_depth (int): the depth of the MLP used to predict
            mask quality
          iou_head_hidden_dim (int): the hidden dimension of the MLP
            used to predict mask quality
        """
        super().__init__()
        self.transformer_dim = transformer_dim
        self.transformer = transformer

        self.num_multimask_outputs = num_multimask_outputs

        self.iou_token = nn.Embedding(1, transformer_dim)
        self.num_mask_tokens = num_multimask_outputs + 1
        self.mask_tokens = nn.Embedding(self.num_mask_tokens, transformer_dim)

        self.output_upscaling = nn.Sequential(
            nn.ConvTranspose2d(transformer_dim, transformer_dim // 4, kernel_size=2, stride=2),
            LayerNorm2d(transformer_dim // 4),
            activation(),
            nn.ConvTranspose2d(transformer_dim // 4, transformer_dim // 8, kernel_size=2, stride=2),
            activation(),
        )
        self.output_hypernetworks_mlps = nn.ModuleList(
            [
                MLP(transformer_dim, transformer_dim, transformer_dim // 8, 3)
                for i in range(self.num_mask_tokens)
            ]
        )

        self.iou_prediction_head = MLP(
            transformer_dim, iou_head_hidden_dim, self.num_mask_tokens, iou_head_depth
        )

    def forward(
        self,
        image_embeddings: torch.Tensor,
        image_pe: torch.Tensor,
        sparse_prompt_embeddings: torch.Tensor,
        dense_prompt_embeddings: torch.Tensor,
        multimask_output: bool,
    ) -> Tuple[torch.Tensor, torch.Tensor]:
        """
        Predict masks given image and prompt embeddings.

        Arguments:
          image_embeddings (torch.Tensor): the embeddings from the image encoder
          image_pe (torch.Tensor): positional encoding with the shape of image_embeddings
          sparse_prompt_embeddings (torch.Tensor): the embeddings of the points and boxes
          dense_prompt_embeddings (torch.Tensor): the embeddings of the mask inputs
          multimask_output (bool): Whether to return multiple masks or a single
            mask.

        Returns:
          torch.Tensor: batched predicted masks
          torch.Tensor: batched predictions of mask quality
        """
        masks, iou_pred = self.predict_masks(
            image_embeddings=image_embeddings,
            image_pe=image_pe,
            sparse_prompt_embeddings=sparse_prompt_embeddings,
            dense_prompt_embeddings=dense_prompt_embeddings,
        )

        # Select the correct mask or masks for output
        if multimask_output:
            mask_slice = slice(1, None)
        else:
            mask_slice = slice(0, 1)
        masks = masks[:, mask_slice, :, :]
        iou_pred = iou_pred[:, mask_slice]

        # Prepare output
        return masks, iou_pred

    def predict_masks(
        self,
        image_embeddings: torch.Tensor,
        image_pe: torch.Tensor,
        sparse_prompt_embeddings: torch.Tensor,
        dense_prompt_embeddings: torch.Tensor,
    ) -> Tuple[torch.Tensor, torch.Tensor]:
        """Predicts masks. See 'forward' for more details."""
        # Concatenate output tokens
        output_tokens = torch.cat([self.iou_token.weight, self.mask_tokens.weight], dim=0)
        output_tokens = output_tokens.unsqueeze(0).expand(sparse_prompt_embeddings.size(0), -1, -1)
        tokens = torch.cat((output_tokens, sparse_prompt_embeddings), dim=1)

        # Expand per-image data in batch direction to be per-mask
        src = torch.repeat_interleave(image_embeddings, tokens.shape[0], dim=0)
        src = src + dense_prompt_embeddings
        pos_src = torch.repeat_interleave(image_pe, tokens.shape[0], dim=0)
        b, c, h, w = src.shape

        # Run the transformer
        hs, src = self.transformer(src, pos_src, tokens)
        iou_token_out = hs[:, 0, :]
        mask_tokens_out = hs[:, 1 : (1 + self.num_mask_tokens), :]

        # Upscale mask embeddings and predict masks using the mask tokens
        src = src.transpose(1, 2).view(b, c, h, w)
        upscaled_embedding = self.output_upscaling(src)
        hyper_in_list: List[torch.Tensor] = []
        for i in range(self.num_mask_tokens):
            hyper_in_list.append(self.output_hypernetworks_mlps[i](mask_tokens_out[:, i, :]))
        hyper_in = torch.stack(hyper_in_list, dim=1)
        b, c, h, w = upscaled_embedding.shape
        masks = (hyper_in @ upscaled_embedding.view(b, c, h * w)).view(b, -1, h, w)

        # Generate mask quality predictions
        iou_pred = self.iou_prediction_head(iou_token_out)

        return masks, iou_pred


# Lightly adapted from
# https://github.com/facebookresearch/MaskFormer/blob/main/mask_former/modeling/transformer/transformer_predictor.py # noqa
class MLP(nn.Module):
    def __init__(
        self,
        input_dim: int,
        hidden_dim: int,
        output_dim: int,
        num_layers: int,
        sigmoid_output: bool = False,
    ) -> None:
        super().__init__()
        self.num_layers = num_layers
        h = [hidden_dim] * (num_layers - 1)
        self.layers = nn.ModuleList(
            nn.Linear(n, k) for n, k in zip([input_dim] + h, h + [output_dim])
        )
        self.sigmoid_output = sigmoid_output

    def forward(self, x):
        for i, layer in enumerate(self.layers):
            x = F.relu(layer(x)) if i < self.num_layers - 1 else layer(x)
        if self.sigmoid_output:
            x = F.sigmoid(x)
        return x

使用?

命令生成掩码：

python scripts/amg.py --checkpoint <path/to/checkpoint> --model-type <model_type> --input <image_or_folder> --output <path/to/output>

print、box输入生成mark

import numpy as np
import torch
import matplotlib.pyplot as plt
import cv2
def show_mask(mask, ax, random_color=False):#接受mask参数
    if random_color:
        color = np.concatenate([np.random.random(3), np.array([0.6])], axis=0)
    else:
        color = np.array([30/255, 144/255, 255/255, 0.6])
    h, w = mask.shape[-2:]
    mask_image = mask.reshape(h, w, 1) * color.reshape(1, 1, -1)
    ax.imshow(mask_image)
    
def show_points(coords, labels, ax, marker_size=375):
    pos_points = coords[labels==1]
    neg_points = coords[labels==0]
    ax.scatter(pos_points[:, 0], pos_points[:, 1], color='green', marker='*', s=marker_size, edgecolor='white', linewidth=1.25)
    ax.scatter(neg_points[:, 0], neg_points[:, 1], color='red', marker='*', s=marker_size, edgecolor='white', linewidth=1.25)   
    
def show_box(box, ax):
    x0, y0 = box[0], box[1]
    w, h = box[2] - box[0], box[3] - box[1]
    ax.add_patch(plt.Rectangle((x0, y0), w, h, edgecolor='green', facecolor=(0,0,0,0), lw=2))    

image = cv2.imread('sam/2.png')
image = cv2.resize(image,None,fx=0.5,fy=0.5)
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)

# plt.figure(figsize=(10,10))
# plt.imshow(image)
# plt.axis('on')\
#加载SAM模型
import sys
sys.path.append("..")
from segment_anything import sam_model_registry, SamPredictor

sam_checkpoint = "sam_vit_b_01ec64.pth"
model_type = "vit_b"

device = "cuda"

sam = sam_model_registry[model_type](checkpoint=sam_checkpoint)
sam.to(device=device)

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

#单点输入生成mask
input_point = np.array([[180, 65]])#点以（x，y）格式输入到模型中，并带有标签1（前景点）或0（背景点）。
input_label = np.array([1])

plt.figure(figsize=(10,10))
plt.imshow(image)
show_points(input_point, input_label, plt.gca())
plt.axis('on')
plt.show()
#使用“SamPredictor.prdict”进行预测。该模型返回掩码（masks）、掩码的分数（scores）以及可传递到下一次预测迭代的低分辨率掩码（logits）  
masks, scores, logits = predictor.predict(#
    point_coords=input_point,
    point_labels=input_label,
    multimask_output=True,
)

print(masks.shape)  # (number_of_masks) x H x W  | output (3, 600, 900)

for i, (mask, score) in enumerate(zip(masks, scores)):
    plt.figure(figsize=(10,10))
    plt.imshow(image)
    show_mask(mask, plt.gca())
    show_points(input_point, input_label, plt.gca())
    plt.title(f"Mask {i+1}, Score: {score:.3f}", fontsize=18)#SAM输出3个掩码，其中“scores”给出了模型对这些掩码质量的估计。
    plt.axis('off')
    plt.show()  

#boxes输入生成mask
#支持将xyxy格式的box作为输入，将框内的主体目标识别出来（类似于实例分割）
    input_box = np.array([82,97,114,130])
masks, _, _ = predictor.predict(
    point_coords=None,
    point_labels=None,
    box=input_box[None, :],
    multimask_output=False,
)
plt.figure(figsize=(10, 10))
plt.imshow(image)
show_mask(masks[0], plt.gca())
show_box(input_box, plt.gca())
plt.axis('off')
plt.show()

自动生成图片掩码

import numpy as np
import torch
import matplotlib.pyplot as plt
import cv2
def show_anns(anns):
    if len(anns) == 0:
        return
    sorted_anns = sorted(anns, key=(lambda x: x['area']), reverse=True)#根据每个标注的面积大小进行降序排列，即面积最大的标注排在最前面
    ax = plt.gca()
    ax.set_autoscale_on(False)

    img = np.ones((sorted_anns[0]['segmentation'].shape[0], sorted_anns[0]['segmentation'].shape[1], 4))#作为最终展示的彩色遮罩图像。
    img[:,:,3] = 0
    for ann in sorted_anns:#获取其分割区域的掩码（m），然后生成一个随机颜色掩码（color_mask），通过将颜色掩码应用于 img 数组的相应位置，来将该标注的区域填充为随机颜色。
        m = ann['segmentation']
        color_mask = np.concatenate([np.random.random(3), [0.35]])
        img[m] = color_mask
    ax.imshow(img)

image = cv2.imread('sam/1.png')
image = cv2.resize(image,None,fx=0.5,fy=0.5)#调用cv2.resize函数将图像缩小到原始大小的一半，缩小系数由fx和fy参数控制
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)#将图像从BGR格式转换为RGB格式

# plt.figure(figsize=(20,20))
# plt.imshow(image)
# plt.axis('off')

# plt.show()
#加载sam
import sys
sys.path.append("..")
from segment_anything import sam_model_registry, SamAutomaticMaskGenerator, SamPredictor

sam_checkpoint = "sam_vit_b_01ec64.pth"
model_type = "vit_b"

device = "cuda"

sam = sam_model_registry[model_type](checkpoint=sam_checkpoint)
sam.to(device=device)
#自动生成采样点对图像进行分割
mask_generator = SamAutomaticMaskGenerator(sam)

masks = mask_generator.generate(image)#使用 mask_generator 对象的 generate 方法生成了一个掩码（mask）列表，用于对原始图像进行分割。

print(len(masks))
print(masks[0].keys())
print(masks[0])

plt.figure(figsize=(16,16))
plt.imshow(image)
show_anns(masks)
plt.axis('off')
plt.show() 

更多的功能使用可以看看官方介绍

notebooks/automatic_mask_generator_example.ipynb

?

使用一个python脚本生成自己想要的mask

import cv2
import os
import numpy as np
from segment_anything import sam_model_registry, SamPredictor

input_dir = 'sam/input'
output_dir = 'sam/output'
crop_mode=True#是否裁剪到最小范围
#alpha_channel是否保留透明通道
print('最好是每加一个点就按w键predict一次')
os.makedirs(output_dir, exist_ok=True)
image_files = [f for f in os.listdir(input_dir) if f.lower().endswith(('.png', '.jpg', '.jpeg','.JPG','.JPEG','.PNG'))]

sam = sam_model_registry["vit_b"](checkpoint="sam_vit_b_01ec64.pth")
_ = sam.to(device="cuda")#注释掉这一行，会用cpu运行，速度会慢很多
predictor = SamPredictor(sam)#SAM预测图像
def mouse_click(event, x, y, flags, param):#鼠标点击事件
    global input_point, input_label, input_stop#全局变量，输入点，
    if not input_stop:#判定标志是否停止输入响应了！
        if event == cv2.EVENT_LBUTTONDOWN :#鼠标左键
            input_point.append([x, y])
            input_label.append(1)#1表示前景点
        elif event == cv2.EVENT_RBUTTONDOWN :#鼠标右键
            input_point.append([x, y])
            input_label.append(0)#0表示背景点
    else:
        if event == cv2.EVENT_LBUTTONDOWN or event == cv2.EVENT_RBUTTONDOWN :#提示添加不了
            print('此时不能添加点,按w退出mask选择模式')


def apply_mask(image, mask, alpha_channel=True):#应用并且响应mask
    if alpha_channel:
        alpha = np.zeros_like(image[..., 0])#制作掩体
        alpha[mask == 1] = 255#兴趣地方标记为1，且为白色
        image = cv2.merge((image[..., 0], image[..., 1], image[..., 2], alpha))#融合图像
    else:
        image = np.where(mask[..., None] == 1, image, 0)
    return image

def apply_color_mask(image, mask, color, color_dark = 0.5):#对掩体进行赋予颜色
    for c in range(3):
        image[:, :, c] = np.where(mask == 1, image[:, :, c] * (1 - color_dark) + color_dark * color[c], image[:, :, c])
    return image

def get_next_filename(base_path, filename):#进行下一个图像
    name, ext = os.path.splitext(filename)
    for i in range(1, 101):
        new_name = f"{name}_{i}{ext}"
        if not os.path.exists(os.path.join(base_path, new_name)):
            return new_name
    return None

def save_masked_image(image, mask, output_dir, filename, crop_mode_):#保存掩盖部分的图像（感兴趣的图像）
    if crop_mode_:
        y, x = np.where(mask)
        y_min, y_max, x_min, x_max = y.min(), y.max(), x.min(), x.max()
        cropped_mask = mask[y_min:y_max+1, x_min:x_max+1]
        cropped_image = image[y_min:y_max+1, x_min:x_max+1]
        masked_image = apply_mask(cropped_image, cropped_mask)
    else:
        masked_image = apply_mask(image, mask)
    filename = filename[:filename.rfind('.')]+'.png'
    new_filename = get_next_filename(output_dir, filename)
    
    if new_filename:
        if masked_image.shape[-1] == 4:
            cv2.imwrite(os.path.join(output_dir, new_filename), masked_image, [cv2.IMWRITE_PNG_COMPRESSION, 9])
        else:
            cv2.imwrite(os.path.join(output_dir, new_filename), masked_image)
        print(f"Saved as {new_filename}")
    else:
        print("Could not save the image. Too many variations exist.")

current_index = 0

cv2.namedWindow("image")
cv2.setMouseCallback("image", mouse_click)
input_point = []
input_label = []
input_stop=False
while True:
    filename = image_files[current_index]
    image_orign = cv2.imread(os.path.join(input_dir, filename))
    image_crop = image_orign.copy()#原图裁剪
    image = cv2.cvtColor(image_orign.copy(), cv2.COLOR_BGR2RGB)#原图色彩转变
    selected_mask = None
    logit_input= None
    while True:
        #print(input_point)
        input_stop=False
        image_display = image_orign.copy()
        display_info = f'{filename} | Press s to save | Press w to predict | Press d to next image | Press a to previous image | Press space to clear | Press q to remove last point '
        cv2.putText(image_display, display_info, (10, 30), cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 255, 255), 2, cv2.LINE_AA)
        for point, label in zip(input_point, input_label):#输入点和输入类型
            color = (0, 255, 0) if label == 1 else (0, 0, 255)
            cv2.circle(image_display, tuple(point), 5, color, -1)
        if selected_mask is not None :
            color = tuple(np.random.randint(0, 256, 3).tolist())
            selected_image = apply_color_mask(image_display,selected_mask, color)

        cv2.imshow("image", image_display)
        key = cv2.waitKey(1)

        if key == ord(" "):
            input_point = []
            input_label = []
            selected_mask = None
            logit_input= None
        elif key == ord("w"):
            input_stop=True
            if len(input_point) > 0 and len(input_label) > 0:
                #todo 预测图像
                predictor.set_image(image)#设置输入图像
                input_point_np = np.array(input_point)#输入暗示点,需要转变array类型才可以输入
                input_label_np = np.array(input_label)#输入暗示点的类型
                #todo 输入暗示信息，将返回masks
                masks, scores, logits= predictor.predict(
                    point_coords=input_point_np,
                    point_labels=input_label_np,
                    mask_input=logit_input[None, :, :] if logit_input is not None else None,
                    multimask_output=True,
                )

                mask_idx=0
                num_masks = len(masks)#masks的数量
                while(1):
                    color = tuple(np.random.randint(0, 256, 3).tolist())#随机列表颜色，就是
                    image_select = image_orign.copy()
                    selected_mask=masks[mask_idx]#选择msks也就是,a,d切换
                    selected_image = apply_color_mask(image_select,selected_mask, color)
                    mask_info = f'Total: {num_masks} | Current: {mask_idx} | Score: {scores[mask_idx]:.2f} | Press w to confirm | Press d to next mask | Press a to previous mask | Press q to remove last point | Press s to save'
                    cv2.putText(selected_image, mask_info, (10, 30), cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 255, 255), 2, cv2.LINE_AA)
                    #todo 显示在当前的图片,
                    cv2.imshow("image", selected_image)

                    key=cv2.waitKey(10)
                    if key == ord('q') and len(input_point)>0:
                        input_point.pop(-1)
                        input_label.pop(-1)
                    elif key == ord('s'):
                        save_masked_image(image_crop, selected_mask, output_dir, filename, crop_mode_=crop_mode)
                    elif key == ord('a') :
                        if mask_idx>0:
                            mask_idx-=1
                        else:
                            mask_idx=num_masks-1
                    elif key == ord('d') :
                        if mask_idx<num_masks-1:
                            mask_idx+=1
                        else:
                            mask_idx=0
                    elif key == ord('w') :
                        break
                    elif key == ord(" "):
                        input_point = []
                        input_label = []
                        selected_mask = None
                        logit_input= None
                        break
                logit_input=logits[mask_idx, :, :]
                print('max score:',np.argmax(scores),' select:',mask_idx)

        elif key == ord('a'):
            current_index = max(0, current_index - 1)
            input_point = []
            input_label = []
            break
        elif key == ord('d'):
            current_index = min(len(image_files) - 1, current_index + 1)
            input_point = []
            input_label = []
            break
        elif key == 27:
            break
        elif key == ord('q') and len(input_point)>0:
            input_point.pop(-1)
            input_label.pop(-1)
        elif key == ord('s') and selected_mask is not None :
            save_masked_image(image_crop, selected_mask, output_dir, filename, crop_mode_=crop_mode)

    if key == 27:
        break

?

当用户运行该程序时，它会读取指定目录中的所有图像文件，并显示第一张图像。用户可以使用鼠标左键和右键在图像上标记前景和背景区域。然后，用户可以按下“w”键，以传递输入点并使用SAM模型进行预测。程序将返回多个掩膜，表示不同的分割结果。

接下来，程序会在选定的掩膜之间循环，并允许用户选择感兴趣的掩膜。用户可以使用鼠标单击来选择不同的掩膜，并且可以使用键盘上的“a”和“d”键切换掩膜。一旦用户选择了一个掩模，他们可以按“w”键来将其保存到输出目录中。

值得注意的是，用户还可以按下“s”键来保存当前正在显示的图像，无论是否选择了掩膜。另外，“space”键可以清除所有输入点和标签，“q”键可以删除最后一个输入点，“a”和“d”键可以在不同的图像之间切换。

最后，该程序提供了一些可配置的参数，如是否裁剪到最小范围（crop_mode）和是否保留透明通道（alpha_channel），以及选择不同的SAM模型。

还有一些大佬集成所有功能在ui

Uminosachi/sd-webui-inpaint-anything: Inpaint Anything extension performs stable diffusion inpainting on a browser UI using masks from Segment Anything. (github.com)