# LOGAN: Unpaired Shape Transform in Latent Overcomplete Space

arXiv: https://arxiv.org/abs/1903.10170 · HTML: https://mlsanity.com/paper/1903.10170/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 25 Mar 2019.

## Built on by (most cited first, 8 of 8)

- [Neural Marching Cubes](https://mlsanity.com/paper/2106.11272/) (2021)
- [ShaDDR: Interactive Example-Based Geometry and Texture Generation via 3D Shape Detailization and Differentiable Rendering](https://mlsanity.com/paper/2306.04889/) (2023)
- [Learning Mesh Representations via Binary Space Partitioning Tree Networks](https://mlsanity.com/paper/2106.14274/) (2021)
- [Multimodal Shape Completion via IMLE](https://mlsanity.com/paper/2106.16237/) (2021)
- [UNIST: Unpaired Neural Implicit Shape Translation Network](https://mlsanity.com/paper/2112.05381/) (2021)
- [CasTGAN: Cascaded Generative Adversarial Network for Realistic Tabular Data Synthesis](https://mlsanity.com/paper/2307.00384/) (2023)
- [Taylor3DNet: Fast 3D Shape Inference With Landmark Points Based Taylor Series](https://mlsanity.com/paper/2201.06845/) (2022)
- [Generalization error property of infoGAN for two-layer neural network](https://mlsanity.com/paper/2310.00443/) (2023)

## Abstract (arXiv)

We introduce LOGAN, a deep neural network aimed at learning general-purpose shape transforms from unpaired domains. The network is trained on two sets of shapes, e.g., tables and chairs, while there is neither a pairing between shapes from the domains as supervision nor any point-wise correspondence between any shapes. Once trained, LOGAN takes a shape from one domain and transforms it into the other. Our network consists of an autoencoder to encode shapes from the two input domains into a common latent space, where the latent codes concatenate multi-scale shape features, resulting in an overcomplete representation. The translator is based on a generative adversarial network (GAN), operating in the latent space, where an adversarial loss enforces cross-domain translation while a feature preservation loss ensures that the right shape features are preserved for a natural shape transform. We conduct ablation studies to validate each of our key network designs and demonstrate superior capabilities in unpaired shape transforms on a variety of examples over baselines and state-of-the-art approaches. We show that LOGAN is able to learn what shape features to preserve during shape translation, either local or non-local, whether content or style, depending solely on the input domains for training.