EpiGRAF: Rethinking training of 3D GANs

Authors: Ivan Skorokhodov, Sergey Tulyakov, Yiqun Wang, Peter Wonka

NeurIPS 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental The resulting model, named Epi GRAF, is an efficient, high-resolution, pure 3D generator, and we test it on four datasets (two introduced in this work) at 2562 and 5122 resolutions. It obtains state-of-the-art image quality, high-fidelity geometry and trains 2.5 faster than the upsampler-based counterparts.
Researcher Affiliation Academia Ivan Skorokhodov Sergey Tulyakov Peter Wonka (...) 7 Acknowledgements We would like to acknowledge support from the SDAIA-KAUST Center of Excellence in Data Science and Artificial Intelligence.
Pseudocode No The paper does not contain any explicitly labeled pseudocode or algorithm blocks.
Open Source Code Yes Code/data/visualizations: https://universome.github.io/epigraf
Open Datasets Yes We employ it for high-resolution 3D-aware image synthesis on four datasets: FFHQ [25], Cats [77], Megascans Plants, and Megascans Food. The last two benchmarks are introduced in our work and contain 360 renderings of photo-realistic scans of different plants and food objects (described in 4). They are much more complex in terms of geometry and are well-suited for assessing the structural limitations of modern 3D-aware generators. (...) Those benchmarks and the rendering code will be made publicly available.
Dataset Splits No The paper does not specify exact percentages or sample counts for training, validation, and test splits. It mentions training data but not explicit splits for reproduction.
Hardware Specification Yes Metrics. We use FID [20] to measure image quality and estimate the training cost for each method in terms of NVidia V100 GPU days needed to complete the training process.
Software Dependencies No The paper mentions software like Style GAN2, Adam, and Py Vista, but does not specify their version numbers to ensure reproducibility of the experimental setup.
Experiment Setup Yes We inherit the training procedure from Style GAN2-ADA [24] with minimal changes. The optimization is performed by Adam [27] with a learning rate of 0.002 and betas of 0 and 0.99 for both G and D. We use β(T) = 0.8 for T = 10000, z N(0, I) and set Rp = 512. D is trained with R1 regularization [37] with γ = 0.05. We train with the overall batch size of 64 for 15M images seen by D for 2562 resolution and 20M for 5122.