Towards Unsupervised Learning of Generative Models for 3D Controllable Image Synthesis
In recent years, Generative Adversarial Networks have achieved impressive results in photorealistic image synthesis. This progress nurtures hopes that one day the classical rendering pipeline can be replaced by efficient models that are learned directly from images. However, current image synthesis models operate in the 2D domain where disentangling 3D properties such as camera viewpoint or object pose is challenging. Furthermore, they lack an interpretable and controllable representation. Our key hypothesis is that the image generation process should be modeled in 3D space as the physical world surrounding us is intrinsically three-dimensional. We define the new task of 3D controllable image synthesis and propose an approach for solving it by reasoning both in 3D space and in the 2D image domain. We demonstrate that our model is able to disentangle latent 3D factors of simple multi-object scenes in an unsupervised fashion from raw images. Compared to pure 2D baselines, it allows for synthesizing scenes that are consistent wrt. changes in viewpoint or object pose. We further evaluate various 3D representations in terms of their usefulness for this challenging task.
Code (1)
Tasks
Image GenerationObjectSimilar Papers 제목 키워드 기반
Survey on Controlable Image Synthesis with Deep Learning
Image synthesis has attracted emerging research interests in academic and industry communities. Deep learning technologies especially the generative models greatly inspired controllable image synthesis approaches and app…
Deep LearningImage GenerationSurveyCGOF++: Controllable 3D Face Synthesis with Conditional Generative Occupancy Fields
Capitalizing on the recent advances in image generation models, existing controllable face image synthesis methods are able to generate high-fidelity images with some levels of controllability, e.g., controlling the shap…
Face GenerationImage GenerationNeRFVeRi3D: Generative Vertex-based Radiance Fields for 3D Controllable Human Image Synthesis
Unsupervised learning of 3D-aware generative adversarial networks has lately made much progress. Some recent work demonstrates promising results of learning human generative models using neural articulated radiance field…
Image GenerationConditional Adversarial Generative Flow for Controllable Image Synthesis
Flow-based generative models show great potential in image synthesis due to its reversible pipeline and exact log-likelihood target, yet it suffers from weak ability for conditional image synthesis, especially for multi-…
Image GenerationStyle Equalization: Unsupervised Learning of Controllable Generative Sequence Models
Controllable generative sequence models with the capability to extract and replicate the style of specific examples enable many applications, including narrating audiobooks in different voices, auto-completing and auto-c…
text-to-speechText to Speech