ACM Transactions on Graphics · 2022 · 129 citations · 25 references
ComputingComputer ScienceComputer VisionNatural Language ProcessingVisual RepresentationBody ImageDeep LearningImage SynthesisText GenerationControllable GenerationGenerative Image SynthesisMixture Of ExpertsText-to-image GenerationText-to-image Translation
Generating high-quality and diverse human images is an important yet challenging task in vision and graphics. However, existing generative models often fall short under the high diversity of clothing shapes and textures. Furthermore, the generation process is even desired to be intuitively controllable for layman users. In this work, we present a text-driven controllable framework, Text2Human, for a high-quality and diverse human generation. We synthesize full-body human images starting from a given human pose with two dedicated steps. 1) With some texts describing the shapes of clothes, the given human pose is first translated to a human parsing map. 2) The final human image is then generated by providing the system with more attributes about the textures of clothes. Specifically, to model the diversity of clothing textures, we build a hierarchical texture-aware codebook that stores multi-scale neural representations for each type of texture. The codebook at the coarse level includes the structural representations of textures, while the codebook at the fine level focuses on the details of textures. To make use of the learned hierarchical codebook to synthesize desired images, a diffusion-based transformer sampler with mixture of experts is firstly employed to sample indices from the coarsest level of the codebook, which then is used to predict the indices of the codebook at finer levels. The predicted indices at different levels are translated to human images by the decoder learned accompanied with hierarchical codebooks. The use of mixture-of-experts allows for the generated image conditioned on the fine-grained text input. The prediction for finer level indices refines the quality of clothing textures. Extensive quantitative and qualitative evaluations demonstrate that our proposed Text2Human framework can generate more diverse and realistic human images compared to state-of-the-art methods. Our project page is https://yumingj.github.io/projects/Text2Human.html. Code and pretrained models are available at https://github.com/yumingj/Text2Human.
25
The Unreasonable Effectiveness of Deep Features as a Perceptual Metric
Richard Zhang, Phillip Isola, Alexei A. Efros et al. · 2018 · 12.4K citations
Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks
Nils Reimers, Iryna Gurevych · 2019 · 10.9K citations · Full text
High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs
Ting-Chun Wang, Ming-Yu Liu, Jun-Yan Zhu et al. · 2018 · 4.4K citations
Semantic Image Synthesis With Spatially-Adaptive Normalization
Taesung Park, Ming-Yu Liu, Ting-Chun Wang et al. · 2019 · 2.8K citations
AttnGAN: Fine-Grained Text to Image Generation with Attentional Generative Adversarial Networks
Tao Xu, Pengchuan Zhang, Qiuyuan Huang et al. · 2018 · 1.9K citations