James Tompkin @jamestompkin.bsky.social · 10/01/2025We prioritize simplicity and performance over functionality. As a minimal baseline, our model does only basic image generation, lacking many features required for downstream tasks. Think of it as DCGAN in 2025 rather than something feature-rich like StyleGAN. We hope this helps further GAN research! 100
James Tompkin @jamestompkin.bsky.social · 10/01/2025Given the well-behaved loss, we move away from the 2015-ish architecture in StyleGAN and implement G and D with a minimalist yet modern architecture---a simplified ConvNeXt. With the two components combined, we obtain a simple GAN baseline that is stable to train and surpasses StyleGAN performance. 110
James Tompkin @jamestompkin.bsky.social · 10/01/2025To further GAN research, we first improve the GAN loss to alleviate mode dropping and non-convergence. This makes GAN optimization sufficiently easy that we can now discard existing GAN tricks w/o training failure. The dependence on outdated GAN-specific architectures is also eliminated. 100
James Tompkin @jamestompkin.bsky.social · 10/01/2025GANs are often criticized for their training instability, and it is often believed that GANs cannot work w/o many engineering tricks. They use outdated network architectures without modern backbone advances. These supposed weaknesses resulted in the abandonment of GAN research in favor of diffusion. 110
James Tompkin @jamestompkin.bsky.social · 10/01/2025Can GANs compete in 2025? In 'The GAN is dead; long live the GAN! A Modern GAN Baseline', we show that a minimalist GAN w/o any tricks can match the performance of EDM with half the size and one-step generation - github.com/brownvc/r3gan - work of Nick Huang, @skylion.bsky.social, Volodymyr Kuleshov 36914
James Tompkin @jamestompkin.bsky.social · 05/12/2024But what if you _really_ like reflections? Local Gaussian Density Mixtures updates lumigraphs by optimizing mixtures of per-view volumes for 🌟maximum shine🌟 #SIGGRAPHAsia2024 xchaowu.github.io/papers/lgdm/... First author Xiuchao Wu is graduating soon and is looking for a job! 0152
