Deep Generative Models for Images and 3D objects
Abstract
Deep generative models have been a driving force in the advancement of artificial intelligence, particularly in the realms of large language models (LLMs) and large-scale text-to-image models. However, the development in other domains, such as 3D object generation, is behind. In this thesis, we study methods to train improved, scalable generative models of images and 3D objects. We also examine how pre-trained generative models in image space can be leveraged for 3D object generation, and how it can be combined with 3D data to build a generative model for 3D objects. First, we introduce a non-parametric structured latent variable model for image generation. It sequentially draws on a latent canvas in a part-by-part fashion, and then decodes the image from the canvas. We show that our method significantly outperforms previous structured image models and is competitive with other generic generative models. Moreover, we show that our model’s inherent compositionality and interpretability bring significant benefits in the low-data learning regime and latent space editing. Second, we build a hierarchical latent point diffusion model for 3D shape generation. It is set up as a variational autoencoder (VAE) with a hierarchical latent space that combines a global shape latent representation with a point-structured latent space. For generation, we train two hierarchical denoising diffusion models (DDMs) in these latent spaces. The hierarchical VAE approach achieves state-of-the-art…
Read the paper
Related studies
- Learning from First-Generation Qualitative Approaches in the IS Discipline: An Evolutionary View and Some Implications for Authors and Evaluators (PART 1/2)
- Commentary on the clinical and preclinical dosage limits of interstitially administered magnetic fluids for therapeutic hyperthermia based on current practice and efficacy models
- Prosthesis: New Technological Opportunities and Innovative Biomedical Devices
- The history and future of neural modeling for cochlear implants
- Ryanodine receptors: waking up from refractoriness
- Novel Transgenic Rabbit Model Sheds Light on the Puzzling Role of Matrix Metalloproteinase-12 in Atherosclerosis
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.