Toward Multimodal Image-to-Image Translation
Authors: Jun-Yan Zhu, Richard Zhang, Deepak Pathak, Trevor Darrell, Alexei A. Efros, Oliver Wang, Eli Shechtman
NeurIPS 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We present a systematic comparison of our method and other variants on both perceptual realism and diversity. We perform a systematic evaluation of these variants by using humans to judge photorealism and a perceptual distance metric [52] to assess output diversity. |
| Researcher Affiliation | Collaboration | Jun-Yan Zhu UC Berkeley Richard Zhang UC Berkeley Deepak Pathak UC Berkeley Trevor Darrell UC Berkeley Alexei A. Efros UC Berkeley Oliver Wang Adobe Research Eli Shechtman Adobe Research |
| Pseudocode | No | The paper does not contain any structured pseudocode or algorithm blocks. |
| Open Source Code | Yes | Code and data are available at https: //github.com/junyanz/Bicycle GAN. The code and additional results are publicly available at https://github.com/junyanz/ Bicycle GAN. |
| Open Datasets | Yes | We test our method on several image-to-image translation problems from prior work, including edges photos [48, 54], Google maps satellite [20], labels images [5], and outdoor night day images [25]. |
| Dataset Splits | No | The paper mentions using 'validation datasets' (Table 1), but does not explicitly provide specific training, validation, and test dataset splits (e.g., percentages, sample counts, or explicit standard split names with their configurations) for reproduction. |
| Hardware Specification | No | The paper acknowledges 'hardware donations from NVIDIA' but does not specify any particular GPU models, CPU types, or other hardware specifications used for running the experiments. |
| Software Dependencies | No | The paper mentions methods and optimizers (e.g., 'Least Squares GANs (LSGANs) variant', 'Adam') and network architectures, but does not provide specific software dependencies with version numbers (e.g., 'PyTorch 1.9' or 'TensorFlow 2.x'). |
| Experiment Setup | Yes | We set the parameters λimage = 10, λlatent = 0.5 and λKL = 0.01 in all our experiments. We train our networks from scratch using Adam [22] with a batch size of 1 and with a learning rate of 0.0002. We choose latent dimension |z| = 8 across all the datasets. |