Space GAN

Space GAN research focuses on improving the controllability and disentanglement of generative adversarial networks (GANs) for image generation and manipulation. Current efforts concentrate on exploring and manipulating GAN latent spaces using techniques like matrix factorization, gradient-based methods, and transformer-based architectures to achieve finer control over generated attributes and mitigate spurious correlations. This work is significant because it addresses limitations in existing GANs, enabling more precise and creative image synthesis and editing across various domains, including facial reenactment and text-to-image generation. Improved control over GANs promises advancements in fields like computer vision, digital art, and virtual reality.

Papers

February 5, 2024

One-shot Neural Face Reenactment via Finding Directions in GAN's Latent Space
Stella Bounareli, Christos Tzelepis, Vasileios Argyriou, Ioannis Patras, Georgios Tzimiropoulos
Latent Space Pre Trained GAN Model Interpretable Direction GAN Latent Space Neural Face Reenactment Space GAN

December 20, 2023

Class Conditional Time Series Generation with Structured Noise Space GAN
Hamidreza Gholamrezaei, Alireza Koochali, Andreas Dengel, Sheraz Ahmed
Generative Model GAN Model C Gan Space GAN Conditional Time Series Generation

October 10, 2023

SC2GAN: Rethinking Entanglement by Self-correcting Correlated GAN Space
Zikun Chen, Han Zhao, Parham Aarabi, Ruowei Jiang
Generative Adversarial Network Latent Space C Gan Quantum Entanglement Latent Direction Space GAN

May 23, 2023

Exploring Semantic Variations in GAN Latent Spaces via Matrix Factorization
Andrey Palaev, Rustam A. Lukmanov, Adil Khan
Latent Space GAN Model Image Editing Matrix Factorization High Quality Image Semantic Variation Space GAN

September 1, 2022

Exploring Gradient-based Multi-directional Controls in GANs
Zikun Chen, Ruowei Jiang, Brendan Duke, Han Zhao, Parham Aarabi
Generative Adversarial Network Latent Space Space GAN

March 31, 2022

TransEditor: Transformer-Based Dual-Space GAN for Highly Controllable Facial Editing
Yanbo Xu, Yueqin Yin, Liming Jiang, Qianyi Wu, Chengyao Zheng, Chen Change Loy, Bo Dai, Wayne Wu
Controllable Editing Latent Space Editing Space GAN Interactive Facial Image Editing

December 2, 2021

FuseDream: Training-Free Text-to-Image Generation with Improved CLIP+GAN Space Optimization
Xingchao Liu, Chengyue Gong, Lemeng Wu, Shujian Zhang, Hao Su, Qiang Liu
GAN Model Text to Image Generation Space GAN