InsetGAN for Full-Body Image Generation

• URL: http://arxiv.org/abs/2203.07293v1
• Date: Mon, 14 Mar 2022 17:01:46 GMT
• Title: InsetGAN for Full-Body Image Generation
• Authors: Anna Fr\"uhst\"uck and Krishna Kumar Singh and Eli Shechtman and Niloy J. Mitra and Peter Wonka and Jingwan Lu
• Abstract summary: We propose a novel method to combine multiple pretrained GANs. One GAN generates a global canvas (e.g., human body) and a set of specialized GANs, or insets, focus on different parts. We demonstrate the setup by combining a full body GAN with a dedicated high-quality face GAN to produce plausible-looking humans.
• Score: 90.71033704904629
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: While GANs can produce photo-realistic images in ideal conditions for certain domains, the generation of full-body human images remains difficult due to the diversity of identities, hairstyles, clothing, and the variance in pose. Instead of modeling this complex domain with a single GAN, we propose a novel method to combine multiple pretrained GANs, where one GAN generates a global canvas (e.g., human body) and a set of specialized GANs, or insets, focus on different parts (e.g., faces, shoes) that can be seamlessly inserted onto the global canvas. We model the problem as jointly exploring the respective latent spaces such that the generated images can be combined, by inserting the parts from the specialized generators onto the global canvas, without introducing seams. We demonstrate the setup by combining a full body GAN with a dedicated high-quality face GAN to produce plausible-looking humans. We evaluate our results with quantitative metrics and user studies.

Related papers

• Controllable Human Image Generation with Personalized Multi-Garments [46.042383679103125]
  BootComp is a novel framework based on text-to-image diffusion models for controllable human image generation with multiple reference garments. We propose a data generation pipeline to construct a large synthetic dataset, consisting of human and multiple-garment pairs. We show the wide-applicability of our framework by adapting it to different types of reference-based generation in the fashion domain.
  arXiv  Detail & Related papers  (2024-11-25T12:37:13Z)
• Do You Guys Want to Dance: Zero-Shot Compositional Human Dance Generation with Multiple Persons [73.21855272778616]
  We introduce a new task, dataset, and evaluation protocol of compositional human dance generation (cHDG) We propose a novel zero-shot framework, dubbed MultiDance-Zero, that can synthesize videos consistent with arbitrary multiple persons and background while precisely following the driving poses.
  arXiv  Detail & Related papers  (2024-01-24T10:44:16Z)
• HyperHuman: Hyper-Realistic Human Generation with Latent Structural Diffusion [114.15397904945185]
  We propose a unified framework, HyperHuman, that generates in-the-wild human images of high realism and diverse layouts. Our model enforces the joint learning of image appearance, spatial relationship, and geometry in a unified network. Our framework yields the state-of-the-art performance, generating hyper-realistic human images under diverse scenarios.
  arXiv  Detail & Related papers  (2023-10-12T17:59:34Z)
• ARAH: Animatable Volume Rendering of Articulated Human SDFs [37.48271522183636]
  We propose a model to create animatable clothed human avatars with detailed geometry that generalize well to out-of-distribution poses. Our algorithm enables efficient point sampling and accurate point canonicalization while generalizing well to unseen poses. Our method achieves state-of-the-art performance on geometry and appearance reconstruction while creating animatable avatars.
  arXiv  Detail & Related papers  (2022-10-18T17:56:59Z)
• MontageGAN: Generation and Assembly of Multiple Components by GANs [11.117357750374035]
  We propose MontageGAN, which is a Generative Adversarial Networks framework for generating multi-layer images. Our method utilized a two-step approach consisting of local GANs and global GAN.
  arXiv  Detail & Related papers  (2022-05-31T07:34:19Z)
• Realistic Full-Body Anonymization with Surface-Guided GANs [7.37907896341367]
  We propose a new anonymization method that generates realistic humans for in-the-wild images. A key part of our design is to guide adversarial nets by dense pixel-to-surface correspondences between an image and a canonical 3D surface. We demonstrate that surface guidance significantly improves image quality and diversity of samples, yielding a highly practical generator.
  arXiv  Detail & Related papers  (2022-01-06T18:57:59Z)
• HumanGAN: A Generative Model of Humans Images [78.6284090004218]
  We present a generative model for images of dressed humans offering control over pose, local body part appearance and garment style. Our model encodes part-based latent appearance vectors in a normalized pose-independent space and warps them to different poses, it preserves body and clothing appearance under varying posture.
  arXiv  Detail & Related papers  (2021-03-11T19:00:38Z)
• PISE: Person Image Synthesis and Editing with Decoupled GAN [64.70360318367943]
  We propose PISE, a novel two-stage generative model for Person Image Synthesis and Editing. For human pose transfer, we first synthesize a human parsing map aligned with the target pose to represent the shape of clothing. To decouple the shape and style of clothing, we propose joint global and local per-region encoding and normalization.
  arXiv  Detail & Related papers  (2021-03-06T04:32:06Z)
• InterFaceGAN: Interpreting the Disentangled Face Representation Learned by GANs [73.27299786083424]
  We propose a framework called InterFaceGAN to interpret the disentangled face representation learned by state-of-the-art GAN models. We first find that GANs learn various semantics in some linear subspaces of the latent space. We then conduct a detailed study on the correlation between different semantics and manage to better disentangle them via subspace projection.
  arXiv  Detail & Related papers  (2020-05-18T18:01:22Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.