DeepI2I: Enabling Deep Hierarchical Image-to-Image Translation by Transferring from GANs

Authors: yaxing wang, Lu Yu, Joost van de Weijer

NeurIPS 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental On many-class image-to-image translation on three datasets (Animal faces, Birds, and Foods) we decrease m FID by at least 35% when compared to the state-of-the-art. Furthermore, we qualitatively and quantitatively demonstrate that transfer learning significantly improves the performance of I2I systems, especially for small datasets.
Researcher Affiliation Collaboration Yaxing Wang, Lu Yu, Joost van de Weijer Computer Vision Center, Universitat Autònoma de Barcelona {yaxing, lu, joost}@cvc.uab.es ... We acknowledge the support from Huawei Kirin Solution.
Pseudocode No The paper describes its approach and training losses in text and mathematical formulas but does not include any pseudocode or algorithm blocks.
Open Source Code Yes Our code and models are made public at: https://github.com/yaxingwang/Deep I2I.
Open Datasets Yes We present our results on four datasets, namely Animal faces [38], Birds [63], Foods [31] and cat2dog [34].
Dataset Splits No We resized all images to 128 128, and split each data into training set (90 %) and test set (10 %).
Hardware Specification No The paper does not provide specific details about the hardware (e.g., GPU models, CPU types, memory) used for running the experiments.
Software Dependencies No The paper does not provide specific version numbers for any software dependencies (e.g., Python, PyTorch, TensorFlow, or other libraries/solvers).
Experiment Setup Yes The training details for all models are in included Suppl. Mat. Sec. A.