models/sam/build_sam.py

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

import urllib.request
# This source code is licensed under the license found in the
# LICENSE file in the root directory of this source tree.
from functools import partial
from pathlib import Path

import torch

from ..common import TwoWayTransformer
from .modeling import ImageEncoderViT, MaskDecoder, PromptEncoder, Sam


def build_sam_vit_h(args = None, checkpoint=None):
    return _build_sam(
        args,
        encoder_embed_dim=1280,
        encoder_depth=32,
        encoder_num_heads=16,
        encoder_global_attn_indexes=[7, 15, 23, 31],
        checkpoint=checkpoint,
    )


build_sam = build_sam_vit_h


def build_sam_vit_l(args, checkpoint=None):
    return _build_sam(
        args,
        encoder_embed_dim=1024,
        encoder_depth=24,
        encoder_num_heads=16,
        encoder_global_attn_indexes=[5, 11, 17, 23],
        checkpoint=checkpoint,
    )


def build_sam_vit_b(args, checkpoint=None):
    return _build_sam(
        args,
        encoder_embed_dim=768,
        encoder_depth=12,
        encoder_num_heads=12,
        encoder_global_attn_indexes=[2, 5, 8, 11],
        checkpoint=checkpoint,
    )


sam_model_registry = {
    "default": build_sam_vit_b,
    "vit_h": build_sam_vit_h,
    "vit_l": build_sam_vit_l,
    "vit_b": build_sam_vit_b,
}


def _build_sam(
    args,
    encoder_embed_dim,
    encoder_depth,
    encoder_num_heads,
    encoder_global_attn_indexes,
    checkpoint=None,
):
    prompt_embed_dim = 256
    image_size = args.image_size
    vit_patch_size = 16
    image_embedding_size = image_size // vit_patch_size
    sam = Sam(
        args,
        image_encoder=ImageEncoderViT(
            args = args,
            depth=encoder_depth,
            embed_dim=encoder_embed_dim,
            img_size=image_size,
            mlp_ratio=4,
            norm_layer=partial(torch.nn.LayerNorm, eps=1e-6),
            num_heads=encoder_num_heads,
            patch_size=vit_patch_size,
            qkv_bias=True,
            use_rel_pos=True,
            # use_rel_pos=False,
            global_attn_indexes=encoder_global_attn_indexes,
            window_size=14,
            out_chans=prompt_embed_dim,
        ),
        prompt_encoder=PromptEncoder(
            embed_dim=prompt_embed_dim,
            image_embedding_size=(image_embedding_size, image_embedding_size),
            input_image_size=(image_size, image_size),
            mask_in_chans=16,
        ),
        mask_decoder=MaskDecoder(
            num_multimask_outputs=args.multimask_output,
            transformer=TwoWayTransformer(
                depth=2,
                embedding_dim=prompt_embed_dim,
                mlp_dim=2048,
                num_heads=8,
            ),
            transformer_dim=prompt_embed_dim,
            iou_head_depth=3,
            iou_head_hidden_dim=256,
        ),
        pixel_mean=[123.675, 116.28, 103.53],
        pixel_std=[58.395, 57.12, 57.375],
    )
    sam.eval()
    checkpoint = Path(checkpoint)
    if checkpoint.name == "sam_vit_b_01ec64.pth" and not checkpoint.exists():
        cmd = input("Download sam_vit_b_01ec64.pth from facebook AI? [y]/n: ")
        if len(cmd) == 0 or cmd.lower() == 'y':
            checkpoint.parent.mkdir(parents=True, exist_ok=True)
            print("Downloading SAM ViT-B checkpoint...")
            urllib.request.urlretrieve(
                "https://dl.fbaipublicfiles.com/segment_anything/sam_vit_b_01ec64.pth",
                checkpoint,
            )
            print(checkpoint.name, " is downloaded!")
    elif checkpoint.name == "sam_vit_h_4b8939.pth" and not checkpoint.exists():
        cmd = input("Download sam_vit_h_4b8939.pth from facebook AI? [y]/n: ")
        if len(cmd) == 0 or cmd.lower() == 'y':
            checkpoint.parent.mkdir(parents=True, exist_ok=True)
            print("Downloading SAM ViT-H checkpoint...")
            urllib.request.urlretrieve(
                "https://dl.fbaipublicfiles.com/segment_anything/sam_vit_h_4b8939.pth",
                checkpoint,
            )
            print(checkpoint.name, " is downloaded!")
    elif checkpoint.name == "sam_vit_l_0b3195.pth" and not checkpoint.exists():
        cmd = input("Download sam_vit_l_0b3195.pth from facebook AI? [y]/n: ")
        if len(cmd) == 0 or cmd.lower() == 'y':
            checkpoint.parent.mkdir(parents=True, exist_ok=True)
            print("Downloading SAM ViT-L checkpoint...")
            urllib.request.urlretrieve(
                "https://dl.fbaipublicfiles.com/segment_anything/sam_vit_l_0b3195.pth",
                checkpoint,
            )
            print(checkpoint.name, " is downloaded!")

        
    if checkpoint is not None:
        with open(checkpoint, "rb") as f:
            state_dict = torch.load(f)
            
        # Create a new state dictionary with only the parameters that exist in the model
        new_state_dict = {k: v for k, v in state_dict.items() if k in sam.state_dict() and sam.state_dict()[k].shape == v.shape}
        sam.load_state_dict(new_state_dict, strict = False)
    return sam