File size: 6,022 Bytes
2df812d |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 |
# 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.
# modified by ziqi-jin
import torch
from functools import partial
from .modeling import ImageEncoderViT, MaskDecoder, PromptEncoder, Sam, TwoWayTransformer
def build_sam_vit_h(checkpoint=None, customized=False):
return _build_sam(
encoder_embed_dim=1280,
encoder_depth=32,
encoder_num_heads=16,
encoder_global_attn_indexes=[7, 15, 23, 31],
checkpoint=checkpoint,
) if not customized else _build_customized_sam(encoder_embed_dim=1280,
encoder_depth=32,
encoder_num_heads=16,
encoder_global_attn_indexes=[7, 15, 23, 31],
checkpoint=checkpoint,)
def build_sam_vit_l(checkpoint=None, customized=False):
return _build_sam(
encoder_embed_dim=1024,
encoder_depth=24,
encoder_num_heads=16,
encoder_global_attn_indexes=[5, 11, 17, 23],
checkpoint=checkpoint,
) if not customized else _build_customized_sam(encoder_embed_dim=1280,
encoder_depth=32,
encoder_num_heads=16,
encoder_global_attn_indexes=[7, 15, 23, 31],
checkpoint=checkpoint,)
def build_sam_vit_b(checkpoint=None, customized=False):
return _build_sam(
encoder_embed_dim=768,
encoder_depth=12,
encoder_num_heads=12,
encoder_global_attn_indexes=[2, 5, 8, 11],
checkpoint=checkpoint,
) if not customized else _build_customized_sam(encoder_embed_dim=1280,
encoder_depth=32,
encoder_num_heads=16,
encoder_global_attn_indexes=[7, 15, 23, 31],
checkpoint=checkpoint,)
sam_model_registry = {
"default": build_sam_vit_h,
"vit_h": build_sam_vit_h,
"vit_l": build_sam_vit_l,
"vit_b": build_sam_vit_b,
}
def _build_sam(
encoder_embed_dim,
encoder_depth,
encoder_num_heads,
encoder_global_attn_indexes,
checkpoint=None,
):
prompt_embed_dim = 256
image_size = 1024
vit_patch_size = 16
image_embedding_size = image_size // vit_patch_size
sam = Sam(
image_encoder=ImageEncoderViT(
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,
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=3,
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],
)
if checkpoint is not None:
with open(checkpoint, "rb") as f:
state_dict = torch.load(f)
sam.load_state_dict(state_dict)
return sam
def _build_customized_sam(encoder_embed_dim,
encoder_depth,
encoder_num_heads,
encoder_global_attn_indexes,
checkpoint=None,):
prompt_embed_dim = 256
image_size = 1024
vit_patch_size = 16
image_embedding_size = image_size // vit_patch_size
image_encoder = ImageEncoderViT(
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,
global_attn_indexes=encoder_global_attn_indexes,
window_size=14,
out_chans=prompt_embed_dim,
)
mask_decoder=MaskDecoder(
num_multimask_outputs=2,
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,
)
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,
)
prompt_encoder.eval()
image_encoder.eval()
if checkpoint is not None:
with open(checkpoint, "rb") as f:
state_dict = torch.load(f)
# only filter the weight of the image_encoder
image_encoder_state_dict = {k.split("image_encoder.")[-1]: v for k, v in state_dict.items() if k.startswith("image_encoder")}
image_encoder.load_state_dict(image_encoder_state_dict)
# # only filter the weight of the mask_decoder
# decoder_state_dict = {k.split("mask_decoder.")[-1]: v for k, v in state_dict.items() if k.startswith("mask_decoder")}
# mask_decoder.load_state_dict(decoder_state_dict)
# only filter the weight of the prompt_encoder
prompt_encoder_state_dict = {k.split("prompt_encoder.")[-1]: v for k, v in state_dict.items() if k.startswith("prompt_encoder")}
prompt_encoder.load_state_dict(prompt_encoder_state_dict)
return image_encoder, prompt_encoder, mask_decoder |