StyleMorph: Disentangled 3D-Aware Image Synthesis with a 3D Morphable StyleGAN
Authors: Eric-Tuan Le, Edward Bartrum, Iasonas Kokkinos
ICLR 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We show competitive synthesis results on 4 datasets (FFHQ faces, AFHQ Cats, Dogs, Wild), while achieving the joint disentanglement of shape, camera pose, object and background texture. and We evaluate our pipeline together with 11 state-of-the-art baselines on the FFHQ (Karras et al., 2019) and AFHQ (Choi et al., 2020) datasets. |
| Researcher Affiliation | Academia | Eric-Tuan Le1 Edward Bartrum1,2 Iasonas Kokkinos1 1 University College London 2 Alan Turing Institute |
| Pseudocode | No | The paper does not contain structured pseudocode or algorithm blocks. |
| Open Source Code | No | We provide more results and videos in the Appendix and will make our code publicly available. |
| Open Datasets | Yes | We evaluate our pipeline together with 11 state-of-the-art baselines on the FFHQ (Karras et al., 2019) and AFHQ (Choi et al., 2020) datasets. |
| Dataset Splits | No | The paper states the datasets used (FFHQ, AFHQ) and their sizes, but does not explicitly provide specific training, validation, or test dataset splits (e.g., percentages or counts) in the main text. |
| Hardware Specification | No | All final models described in this research were trained using the Baskerville Tier 2 HPC service (https://www.baskerville.ac.uk/). This does not provide specific hardware models like GPUs or CPUs. |
| Software Dependencies | No | The paper does not provide specific ancillary software details with version numbers. |
| Experiment Setup | No | The paper mentions training strategies, loss functions, and regularization methods, but does not provide specific experimental setup details such as concrete hyperparameter values (e.g., learning rate, batch size, number of epochs) in the main text. |