# Appearance and Pose-Conditioned Human Image Generation using Deformable GANs

arXiv: https://arxiv.org/abs/1905.00007 · HTML: https://mlsanity.com/paper/1905.00007/

## Key facts

- Cited by 24 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 30 Apr 2019.

## Built on by (most cited first, 24 of 24)

- [MagicPose: Realistic Human Poses and Facial Expressions Retargeting with Identity-aware Diffusion](https://mlsanity.com/paper/2311.12052/) (2023)
- [HyperHuman: Hyper-Realistic Human Generation with Latent Structural Diffusion](https://mlsanity.com/paper/2310.08579/) (2023)
- [Human4DiT: 360-degree Human Video Generation with 4D Diffusion Transformer](https://mlsanity.com/paper/2405.17405/) (2024)
- [Neural Re-Rendering of Humans from a Single Image](https://mlsanity.com/paper/2101.04104/) (2021)
- [Style and Pose Control for Image Synthesis of Humans from a Single Monocular View](https://mlsanity.com/paper/2102.11263/) (2021)
- [HumanGAN: A Generative Model of Humans Images](https://mlsanity.com/paper/2103.06902/) (2021)
- [InsetGAN for Full-Body Image Generation](https://mlsanity.com/paper/2203.07293/) (2022)
- [Smoothing the Disentangled Latent Style Space for Unsupervised Image-to-Image Translation](https://mlsanity.com/paper/2106.09016/) (2021)
- [3D-Aware Semantic-Guided Generative Model for Human Synthesis](https://mlsanity.com/paper/2112.01422/) (2021)
- [The Center of Attention: Center-Keypoint Grouping via Attention for Multi-Person Pose Estimation](https://mlsanity.com/paper/2110.05132/) (2021)
- [Learning Realistic Human Reposing using Cyclic Self-Supervision with 3D Shape, Pose, and Appearance Consistency](https://mlsanity.com/paper/2110.05458/) (2021)
- [Human Image Generation: A Comprehensive Survey](https://mlsanity.com/paper/2212.08896/) (2022)
- [Cross-identity Video Motion Retargeting with Joint Transformation and Synthesis](https://mlsanity.com/paper/2210.01559/) (2022)
- [Interspatial Attention for Efficient 4D Human Video Generation](https://mlsanity.com/paper/2505.15800/) (2025)
- [DiffBody: Human Body Restoration by Imagining with Generative Diffusion Prior](https://mlsanity.com/paper/2404.03642/) (2024)
- [Multi-focal Conditioned Latent Diffusion for Person Image Synthesis](https://mlsanity.com/paper/2503.15686/) (2025)
- [Unsupervised High-Resolution Portrait Gaze Correction and Animation](https://mlsanity.com/paper/2207.00256/) (2022)
- [Total Selfie: Generating Full-Body Selfies](https://mlsanity.com/paper/2308.14740/) (2023)
- [Multi-Domain Image-to-Image Translation with Adaptive Inference Graph](https://mlsanity.com/paper/2101.03806/) (2021)
- [Controllable Person Image Synthesis with Spatially-Adaptive Warped Normalization](https://mlsanity.com/paper/2105.14739/) (2021)
- [Synthesizing Environment-Specific People in Photographs](https://mlsanity.com/paper/2312.14579/) (2023)
- [MotionWeaver: Holistic 4D-Anchored Framework for Multi-Humanoid Image Animation](https://mlsanity.com/paper/2602.13326/) (2026)
- [Multi-View Image-to-Image Translation Supervised by 3D Pose](https://mlsanity.com/paper/2104.05779/) (2021)
- [Ebola Optimization Search Algorithm (EOSA): A new metaheuristic algorithm based on the propagation model of Ebola virus disease](https://mlsanity.com/paper/2106.01416/) (2021)

## Abstract (arXiv)

In this paper, we address the problem of generating person images conditioned on both pose and appearance information. Specifically, given an image xa of a person and a target pose P(xb), extracted from a different image xb, we synthesize a new image of that person in pose P(xb), while preserving the visual details in xa. In order to deal with pixel-to-pixel misalignments caused by the pose differences between P(xa) and P(xb), we introduce deformable skip connections in the generator of our Generative Adversarial Network. Moreover, a nearest-neighbour loss is proposed instead of the common L1 and L2 losses in order to match the details of the generated image with the target image. Quantitative and qualitative results, using common datasets and protocols recently proposed for this task, show that our approach is competitive with respect to the state of the art. Moreover, we conduct an extensive evaluation using off-the-shell person re-identification (Re-ID) systems trained with person-generation based augmented data, which is one of the main important applications for this task. Our experiments show that our Deformable GANs can significantly boost the Re-ID accuracy and are even better than data-augmentation methods specifically trained using Re-ID losses.