您好,登錄后才能下訂單哦!
這篇文章主要介紹了CoordConv如何實現(xiàn)卷積加上坐標(biāo)的相關(guān)知識,內(nèi)容詳細(xì)易懂,操作簡單快捷,具有一定借鑒價值,相信大家閱讀完這篇CoordConv如何實現(xiàn)卷積加上坐標(biāo)文章都會有所收獲,下面我們一起來看看吧。
這是一篇考古的論文復(fù)現(xiàn)項目,在2018年Uber團(tuán)隊提出這個CoordConv模塊的時候有很多文章對其進(jìn)行批評,認(rèn)為這個不值得發(fā)布一篇論文,但是現(xiàn)在重新看一下這個idea,同時再對比一下目前Transformer中提出的位置編碼(Position Encoding),你就會感概歷史是個圈,在角點卷積中,為卷積添加兩個坐標(biāo)編碼實際上與Transformer中提出的位置編碼是同樣的道理。 眾所周知,深度學(xué)習(xí)里的卷積運算是具有平移等變性的,這樣可以在圖像的不同位置共享統(tǒng)一的卷積核參數(shù),但是這樣卷積學(xué)習(xí)過程中是不能感知當(dāng)前特征在圖像中的坐標(biāo)的,論文中的實驗證明如下圖所示。通過該實驗,作者證明了傳統(tǒng)卷積在卷積核進(jìn)行局部運算時,僅僅能感受到局部信息,并且是無法感受到位置信息的。CoordConv就是通過在卷積的輸入特征圖中新增對應(yīng)的通道來表征特征圖像素點的坐標(biāo),讓卷積學(xué)習(xí)過程中能夠一定程度感知坐標(biāo)來提升檢測精度。
傳統(tǒng)卷積無法將空間表示轉(zhuǎn)換成笛卡爾空間中的坐標(biāo)和one-hot像素空間中的坐標(biāo)。卷積是等變的,也就是說當(dāng)每個過濾器應(yīng)用到輸入上時,它不知道每個過濾器在哪。我們可以幫助卷積,讓它知道過濾器的位置。這一過程需要在輸入上添加兩個通道實現(xiàn),一個在i坐標(biāo),另一個在j坐標(biāo)。通過上面的添加坐標(biāo)的操作,我們可以的出一種新的卷積結(jié)構(gòu)--CoordConv,其結(jié)構(gòu)如下圖所示:
本部分根據(jù)CoordConv論文并參考飛槳的官方實現(xiàn)完成CoordConv的復(fù)現(xiàn)。
import paddle import paddle.nn as nn import paddle.nn.functional as F from paddle import ParamAttr from paddle.regularizer import L2Decay from paddle.nn import AvgPool2D, Conv2D
首先繼承nn.Layer基類,其次使用paddle.arange
定義gx``gy
兩個坐標(biāo),并且停止它們的梯度反傳gx.stop_gradient = True
,最后將它們concat到一起送入卷積即可。
class CoordConv(nn.Layer): def __init__(self, in_channels, out_channels, kernel_size, stride, padding): super(CoordConv, self).__init__() self.conv = Conv2D( in_channels + 2, out_channels , kernel_size , stride , padding) def forward(self, x): b = x.shape[0] h = x.shape[2] w = x.shape[3] gx = paddle.arange(w, dtype='float32') / (w - 1.) * 2.0 - 1. gx = gx.reshape([1, 1, 1, w]).expand([b, 1, h, w]) gx.stop_gradient = True gy = paddle.arange(h, dtype='float32') / (h - 1.) * 2.0 - 1. gy = gy.reshape([1, 1, h, 1]).expand([b, 1, h, w]) gy.stop_gradient = True y = paddle.concat([x, gx, gy], axis=1) y = self.conv(y) return y
class dcn2(paddle.nn.Layer): def __init__(self, num_classes=1): super(dcn2, self).__init__() self.conv1 = paddle.nn.Conv2D(in_channels=3, out_channels=32, kernel_size=(3, 3), stride=1, padding = 1) self.conv2 = paddle.nn.Conv2D(in_channels=32, out_channels=64, kernel_size=(3,3), stride=2, padding = 0) self.conv3 = paddle.nn.Conv2D(in_channels=64, out_channels=64, kernel_size=(3,3), stride=2, padding = 0) self.offsets = paddle.nn.Conv2D(64, 18, kernel_size=3, stride=2, padding=1) self.mask = paddle.nn.Conv2D(64, 9, kernel_size=3, stride=2, padding=1) self.conv4 = CoordConv(64, 64, (3,3), 2, 1) self.flatten = paddle.nn.Flatten() self.linear1 = paddle.nn.Linear(in_features=1024, out_features=64) self.linear2 = paddle.nn.Linear(in_features=64, out_features=num_classes) def forward(self, x): x = self.conv1(x) x = F.relu(x) x = self.conv2(x) x = F.relu(x) x = self.conv3(x) x = F.relu(x) x = self.conv4(x) x = F.relu(x) x = self.flatten(x) x = self.linear1(x) x = F.relu(x) x = self.linear2(x) return x
cnn3 = dcn2() model3 = paddle.Model(cnn3) model3.summary((64, 3, 32, 32))
--------------------------------------------------------------------------- Layer (type) Input Shape Output Shape Param # =========================================================================== Conv2D-26 [[64, 3, 32, 32]] [64, 32, 32, 32] 896 Conv2D-27 [[64, 32, 32, 32]] [64, 64, 15, 15] 18,496 Conv2D-28 [[64, 64, 15, 15]] [64, 64, 7, 7] 36,928 Conv2D-31 [[64, 66, 7, 7]] [64, 64, 4, 4] 38,080 CoordConv-4 [[64, 64, 7, 7]] [64, 64, 4, 4] 0 Flatten-1 [[64, 64, 4, 4]] [64, 1024] 0 Linear-1 [[64, 1024]] [64, 64] 65,600 Linear-2 [[64, 64]] [64, 1] 65 =========================================================================== Total params: 160,065 Trainable params: 160,065 Non-trainable params: 0 --------------------------------------------------------------------------- Input size (MB): 0.75 Forward/backward pass size (MB): 26.09 Params size (MB): 0.61 Estimated Total Size (MB): 27.45 --------------------------------------------------------------------------- {'total_params': 160065, 'trainable_params': 160065}
class MyNet(paddle.nn.Layer): def __init__(self, num_classes=1): super(MyNet, self).__init__() self.conv1 = paddle.nn.Conv2D(in_channels=3, out_channels=32, kernel_size=(3, 3), stride=1, padding = 1) self.conv2 = paddle.nn.Conv2D(in_channels=32, out_channels=64, kernel_size=(3,3), stride=2, padding = 0) self.conv3 = paddle.nn.Conv2D(in_channels=64, out_channels=64, kernel_size=(3,3), stride=2, padding = 0) self.conv4 = paddle.nn.Conv2D(in_channels=64, out_channels=64, kernel_size=(3,3), stride=2, padding = 1) self.flatten = paddle.nn.Flatten() self.linear1 = paddle.nn.Linear(in_features=1024, out_features=64) self.linear2 = paddle.nn.Linear(in_features=64, out_features=num_classes) def forward(self, x): x = self.conv1(x) x = F.relu(x) x = self.conv2(x) x = F.relu(x) x = self.conv3(x) x = F.relu(x) x = self.conv4(x) x = F.relu(x) x = self.flatten(x) x = self.linear1(x) x = F.relu(x) x = self.linear2(x) return x
# 可視化模型 cnn1 = MyNet() model1 = paddle.Model(cnn1) model1.summary((64, 3, 32, 32))
--------------------------------------------------------------------------- Layer (type) Input Shape Output Shape Param # =========================================================================== Conv2D-1 [[64, 3, 32, 32]] [64, 32, 32, 32] 896 Conv2D-2 [[64, 32, 32, 32]] [64, 64, 15, 15] 18,496 Conv2D-3 [[64, 64, 15, 15]] [64, 64, 7, 7] 36,928 Conv2D-4 [[64, 64, 7, 7]] [64, 64, 4, 4] 36,928 Flatten-1 [[64, 64, 4, 4]] [64, 1024] 0 Linear-1 [[64, 1024]] [64, 64] 65,600 Linear-2 [[64, 64]] [64, 1] 65 =========================================================================== Total params: 158,913 Trainable params: 158,913 Non-trainable params: 0 --------------------------------------------------------------------------- Input size (MB): 0.75 Forward/backward pass size (MB): 25.59 Params size (MB): 0.61 Estimated Total Size (MB): 26.95 --------------------------------------------------------------------------- {'total_params': 158913, 'trainable_params': 158913}
關(guān)于“CoordConv如何實現(xiàn)卷積加上坐標(biāo)”這篇文章的內(nèi)容就介紹到這里,感謝各位的閱讀!相信大家對“CoordConv如何實現(xiàn)卷積加上坐標(biāo)”知識都有一定的了解,大家如果還想學(xué)習(xí)更多知識,歡迎關(guān)注億速云行業(yè)資訊頻道。
免責(zé)聲明:本站發(fā)布的內(nèi)容(圖片、視頻和文字)以原創(chuàng)、轉(zhuǎn)載和分享為主,文章觀點不代表本網(wǎng)站立場,如果涉及侵權(quán)請聯(lián)系站長郵箱:is@yisu.com進(jìn)行舉報,并提供相關(guān)證據(jù),一經(jīng)查實,將立刻刪除涉嫌侵權(quán)內(nèi)容。