File size: 17,613 Bytes
1715fda |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 |
# -*- coding: utf-8 -*-
# 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
from .transformer import TwoWayTransformer
class Classifier(nn.Module):
def __init__(self, in_dim, hid_dim=None, out_dim=None, act=nn.GELU, drop=0.):
super().__init__()
out_dim = out_dim or in_dim
hid_dim = hid_dim or in_dim
self.fc1 = nn.Linear(in_dim, hid_dim)
self.act = act()
self.fc2 = nn.Linear(hid_dim, out_dim)
self.drop = nn.Dropout(drop)
def forward(self, x):
x = self.fc1(x)
x = self.act(x)
x = self.drop(x)
x = self.fc2(x)
return x
class Block(nn.Module):
def __init__(self, in_channels, out_channels, i_downsample=None, stride=1):
super(Block, self).__init__()
self.conv1 = nn.Conv2d(in_channels, out_channels, kernel_size=3, padding=1, stride=stride, bias=False)
self.batch_norm1 = nn.BatchNorm2d(out_channels)
self.conv2 = nn.Conv2d(out_channels, out_channels, kernel_size=3, padding=1, stride=stride, bias=False)
self.i_downsample = i_downsample
self.stride = stride
self.relu = nn.LeakyReLU(negative_slope=0.1, inplace=True)
def forward(self, x):
identity = x.clone()
x = self.relu(self.batch_norm1(self.conv1(x)))
x = self.conv2(x)
if self.i_downsample is not None:
identity = self.i_downsample(identity)
x += identity
return x
class MaskDecoder(nn.Module):
def __init__(
self,
*,
transformer_dim: int,
transformer: nn.Module,
modality,
contents,
num_multimask_outputs: int = 3,
activation: Type[nn.Module] = nn.GELU,
iou_head_depth: int = 3,
iou_head_hidden_dim: int = 256,
category_num = 11
) -> 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.category_num = category_num
self.modality = modality
self.contents = contents
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.convs = Block(transformer_dim, transformer_dim)
self.w_lin = nn.Linear(transformer_dim, transformer_dim)
self.b_lin = nn.Linear(transformer_dim, 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
)
self.category_prediction_head = Classifier(
transformer_dim, transformer_dim//4, category_num
)
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, category_pred, clip_tokens_out, image_tokens_out = 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, category_pred, clip_tokens_out, image_tokens_out
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
if image_embeddings.shape[0] != tokens.shape[0]:
src = torch.repeat_interleave(image_embeddings, tokens.shape[0], dim=0)
else:
src = image_embeddings
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)
if self.contents:
clip_tokens_out = tokens[:,-2,:]
image_tokens_out = F.adaptive_avg_pool2d(dense_prompt_embeddings, output_size=(1, 1)).squeeze(-1).squeeze(-1)
clip_new_out = hs[:,-2,:].unsqueeze(-1).unsqueeze(-1)
src = dense_prompt_embeddings+src+clip_new_out
src = self.convs(src)
else:
clip_tokens_out = None
image_tokens_out = None
if self.modality:
category_tokens_out = hs[:,-1,:]
wc = self.w_lin(category_tokens_out).unsqueeze(-1).unsqueeze(-1)
bc = self.b_lin(category_tokens_out).unsqueeze(-1).unsqueeze(-1)
src = wc*src+bc+src
category_pred = self.category_prediction_head(category_tokens_out)
else:
category_pred = None
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, category_pred, clip_tokens_out, image_tokens_out
# 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
class MaskDecoder_F4(nn.Module):
def __init__(
self,
*,
transformer_dim: int,
transformer: nn.Module,
modality,
contents,
num_multimask_outputs: int = 3,
activation: Type[nn.Module] = nn.GELU,
iou_head_depth: int = 3,
iou_head_hidden_dim: int = 256,
category_num = 11
) -> 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.category_num = category_num
self.modality = modality
self.contents = contents
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.convs = Block(transformer_dim, transformer_dim)
self.conv1 = nn.Conv2d(transformer_dim*2, transformer_dim, 1)
self.c_conv = Block(transformer_dim, transformer_dim)
self.w_lin = nn.Linear(transformer_dim, transformer_dim)
self.b_lin = nn.Linear(transformer_dim, transformer_dim)
self.m_conv = Block(transformer_dim, 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
)
# self.category_prediction_head = Classifier(
# transformer_dim, transformer_dim//4, category_num
# )
self.category_prediction_head = Classifier(
transformer_dim, transformer_dim//4, category_num
)
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, category_pred, clip_tokens_out, image_tokens_out = 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, category_pred, clip_tokens_out, image_tokens_out
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)
m_token = tokens[:,-1,:]
# Expand per-image data in batch direction to be per-mask
if image_embeddings.shape[0] != tokens.shape[0]:
src = torch.repeat_interleave(image_embeddings, tokens.shape[0], dim=0)
else:
src = image_embeddings
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)
if self.modality:
category_tokens_out = hs[:,-1,:]
wc = self.w_lin(category_tokens_out).unsqueeze(-1).unsqueeze(-1)
bc = self.b_lin(category_tokens_out).unsqueeze(-1).unsqueeze(-1)
src_m = wc*src+bc+src
m_info = wc.squeeze(-1).squeeze(-1)+bc.squeeze(-1).squeeze(-1)+category_tokens_out
category_pred = self.category_prediction_head(m_info)
src_m = self.m_conv(src_m)
else:
category_pred = None
if self.contents:
clip_tokens_out = tokens[:,-2,:]
image_tokens_out = F.adaptive_avg_pool2d(dense_prompt_embeddings, output_size=(1, 1)).squeeze(-1).squeeze(-1)
clip_new_out = hs[:,-2,:].unsqueeze(-1).unsqueeze(-1)
src_vp = dense_prompt_embeddings+src+clip_new_out
src_vp = self.convs(src_vp)
else:
clip_tokens_out = None
image_tokens_out = None
if self.contents and self.modality:
src = torch.cat((src_m, src_vp), dim=1)
src = self.conv1(src)
src = self.c_conv(src)
elif self.contents:
src = src_vp
elif self.modality:
src = src_m
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, category_pred, clip_tokens_out, image_tokens_out
|