ICML 2021 accepted papers on Image generation
22 of the 1,184 papers accepted at ICML 2021 (1.9%) are on Image generation, compared with 1.6% at ICML 2020. They fall into 7 topics.
See ICML 2021 in the interactive map
Submitting to ICML? Checklist →
Working on something in this area? Describe your idea in scime atlas to see the closest papers by meaning and how crowded the spot has become.
Topics and papers
Autoregressive and latent generation · 10 papers
- Meta-StyleSpeech : Multi-Speaker Adaptive Text-to-Speech Generation (45 citations)
- Improved Contrastive Divergence Training of Energy-Based Models (26 citations)
- BANG: Bridging Autoregressive and Non-autoregressive Generation with Large Scale Pretraining (24 citations)
- Generative Adversarial Transformers (17 citations)
- Autoencoder Image Interpolation by Shaping the Latent Space (9 citations)
- SketchEmbedNet: Learning Novel Concepts by Imitating Drawings (9 citations)
- Generating images with sparse representations (8 citations)
- Testing DNN-based Autonomous Driving Systems under Critical Environmental Conditions (7 citations)
- Prior Image-Constrained Reconstruction using Style-Based Generative Models (5 citations)
- PixelTransformer: Sample Conditioned Signal Generation
GANs · 4 papers
- Pareto GAN: Extending the Representational Power of GANs to Heavy-Tailed Distributions (2 citations)
- Generative Adversarial Networks for Markovian Temporal Dynamics: Stochastic Continuous Data Generation (1 citation)
- Uncertainty Principles of Encoding GANs (1 citation)
- Understanding Noise Injection in GANs
Image restoration · 2 papers
Fast diffusion sampling · 2 papers
- Improved Denoising Diffusion Probabilistic Models (411 citations)
- Grad-TTS: A Diffusion Probabilistic Model for Text-to-Speech (43 citations)
Editing and style transfer · 2 papers
Super-resolution · 1 paper
Text-to-image diffusion · 1 paper
- Zero-Shot Text-to-Image Generation (1,121 citations)
