DeepAI AI Chat
Log In Sign Up

Semantic Bottleneck Scene Generation

by   Samaneh Azadi, et al.
berkeley college

Coupling the high-fidelity generation capabilities of label-conditional image synthesis methods with the flexibility of unconditional generative models, we propose a semantic bottleneck GAN model for unconditional synthesis of complex scenes. We assume pixel-wise segmentation labels are available during training and use them to learn the scene structure. During inference, our model first synthesizes a realistic segmentation layout from scratch, then synthesizes a realistic scene conditioned on that layout. For the former, we use an unconditional progressive segmentation generation network that captures the distribution of realistic semantic scene layouts. For the latter, we use a conditional segmentation-to-image synthesis network that captures the distribution of photo-realistic images conditioned on the semantic layout. When trained end-to-end, the resulting model outperforms state-of-the-art generative models in unsupervised image synthesis on two challenging domains in terms of the Frechet Inception Distance and user-study evaluations. Moreover, we demonstrate the generated segmentation maps can be used as additional training data to strongly improve recent segmentation-to-image synthesis networks.


page 5

page 7

page 8

page 11

page 12

page 13

page 14

page 15


Semantic Palette: Guiding Scene Generation with Class Proportions

Despite the recent progress of generative adversarial networks (GANs) at...

Full-Glow: Fully conditional Glow for more realistic image generation

Autonomous agents, such as driverless cars, require large amounts of lab...

Unifying conditional and unconditional semantic image synthesis with OCO-GAN

Generative image models have been extensively studied in recent years. I...

End-to-End Optimization of Scene Layout

We propose an end-to-end variational generative model for scene layout s...

Interactive Image Synthesis with Panoptic Layout Generation

Interactive image synthesis from user-guided input is a challenging task...

Photographic Image Synthesis with Cascaded Refinement Networks

We present an approach to synthesizing photographic images conditioned o...

Language Conditioned Traffic Generation

Simulation forms the backbone of modern self-driving development. Simula...

Code Repositories


Semantic Bottleneck Scene Generation

view repo