FDA: Fourier Domain Adaptation for Semantic Segmentation

We describe a simple method for unsupervised domain adaptation, whereby the discrepancy between the source and target distributions is reduced by swapping the low-frequency spectrum of one with the other. We illustrate the method in semantic segmentation, where densely annotated images are aplenty in one domain (synthetic data), but difficult to obtain in another (real images)... (read more)

PDF Abstract CVPR 2020 PDF CVPR 2020 Abstract
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
Domain Adaptation SYNTHIA-to-Cityscapes FDA (VGG-16) mIoU 40.5 # 12

Methods used in the Paper


METHOD TYPE
Batch Normalization
Normalization
Spatial Pyramid Pooling
Pooling Operations
Average Pooling
Pooling Operations
1x1 Convolution
Convolutions
Residual Connection
Skip Connections
FCN
Semantic Segmentation Models
Global Average Pooling
Pooling Operations
Bottleneck Residual Block
Skip Connection Blocks
Residual Block
Skip Connection Blocks
Kaiming Initialization
Initialization
ResNet
Convolutional Neural Networks
ASPP
Semantic Segmentation Modules
Dilated Convolution
Convolutions
Feedforward Network
Feedforward Networks
CRF
Structured Prediction
DeepLabv2
Semantic Segmentation Models
Dropout
Regularization
ReLU
Activation Functions
Max Pooling
Pooling Operations
Convolution
Convolutions
Dense Connections
Feedforward Networks
VGG
Convolutional Neural Networks
Softmax
Output Functions
SGD
Stochastic Optimization
Adam
Stochastic Optimization