Diffusion with Forward Models: Solving Stochastic Inverse Problems Without Direct Supervision

Published: 21 Sept 2023, Last Modified: 16 Jan 2024NeurIPS 2023 spotlightEveryoneRevisionsBibTeX
Keywords: 3D generative models, neural rendering, neural scene representations, NeRF, diffusion models, differentiable rendering, inverse graphics, inverse problems
TL;DR: We propose a novel class of diffusion models that learn to sample from distributions of signals that are never directly observed, e.g., our model learns to sample 3D scenes by only training on 2D image observations.
Abstract: Denoising diffusion models are a powerful type of generative models used to capture complex distributions of real-world signals. However, their applicability is limited to scenarios where training samples are readily available, which is not always the case in real-world applications. For example, in inverse graphics, the goal is to generate samples from a distribution of 3D scenes that align with a given image, but ground-truth 3D scenes are unavailable and only 2D images are accessible. To address this limitation, we propose a novel class of denoising diffusion probabilistic models that learn to sample from distributions of signals that are never directly observed. Instead, these signals are measured indirectly through a known differentiable forward model, which produces partial observations of the unknown signal. Our approach involves integrating the forward model directly into the denoising process. A key contribution of our work is the integration of a differentiable forward model into the denoising process. This integration effectively connects the generative modeling of observations with the generative modeling of the underlying signals, allowing for end-to-end training of a conditional generative model over signals. During inference, our approach enables sampling from the distribution of underlying signals that are consistent with a given partial observation. We demonstrate the effectiveness of our method on three challenging computer vision tasks. For instance, in the context of inverse graphics, our model enables direct sampling from the distribution of 3D scenes that align with a single 2D input image.
Supplementary Material: pdf
Submission Number: 8153
Loading

OpenReview is a long-term project to advance science through improved peer review with legal nonprofit status. We gratefully acknowledge the support of the OpenReview Sponsors. © 2025 OpenReview