Big GANs Are Watching You: Towards Unsupervised Object Segmentation with Off-the-Shelf Generative Models

Since collecting pixel-level groundtruth data is expensive, unsupervised visual understanding problems are currently an active research topic. In particular, several recent methods based on generative models have achieved promising results for object segmentation and saliency detection... (read more)

PDF Abstract

Results from the Paper


  Submit results from this paper to get state-of-the-art GitHub badges and help the community compare results to other papers.

Methods used in the Paper


METHOD TYPE
Reversible Residual Block
Skip Connection Blocks
Softmax
Output Functions
Dot-Product Attention
Attention Mechanisms
1x1 Convolution
Convolutions
TTUR
Optimization
SAGAN Self-Attention Module
Attention Modules
Off-Diagonal Orthogonal Regularization
Regularization
Spectral Normalization
Normalization
GAN Hinge Loss
Loss Functions
Early Stopping
Regularization
Dense Connections
Feedforward Networks
Average Pooling
Pooling Operations
Convolution
Convolutions
Non-Local Operation
Image Feature Extractors
Pointwise Convolution
Convolutions
Non-Local Block
Image Model Blocks
Truncation Trick
Latent Variable Sampling
Projection Discriminator
Discriminators
Residual Connection
Skip Connections
CReLU
Activation Functions
RevNet
Convolutional Neural Networks
ReLU
Activation Functions
Batch Normalization
Normalization
Adam
Stochastic Optimization
Feedforward Network
Feedforward Networks
Conditional Batch Normalization
Normalization
SAGAN
Generative Adversarial Networks
BigGAN
Generative Models
Residual Block
Skip Connection Blocks
Global Average Pooling
Pooling Operations
Linear Layer
Feedforward Networks
Softplus
Activation Functions
BigBiGAN
Generative Models