Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

DiffDreamer: Towards Consistent Unsupervised Single-view Scene Extrapolation with Conditional Diffusion Models

About

Scene extrapolation -- the idea of generating novel views by flying into a given image -- is a promising, yet challenging task. For each predicted frame, a joint inpainting and 3D refinement problem has to be solved, which is ill posed and includes a high level of ambiguity. Moreover, training data for long-range scenes is difficult to obtain and usually lacks sufficient views to infer accurate camera poses. We introduce DiffDreamer, an unsupervised framework capable of synthesizing novel views depicting a long camera trajectory while training solely on internet-collected images of nature scenes. Utilizing the stochastic nature of the guided denoising steps, we train the diffusion models to refine projected RGBD images but condition the denoising steps on multiple past and future frames for inference. We demonstrate that image-conditioned diffusion models can effectively perform long-range scene extrapolation while preserving consistency significantly better than prior GAN-based methods. DiffDreamer is a powerful and efficient solution for scene extrapolation, producing impressive results despite limited supervision. Project page: https://primecai.github.io/diffdreamer.

Shengqu Cai, Eric Ryan Chan, Songyou Peng, Mohamad Shahbazi, Anton Obukhov, Luc Van Gool, Gordon Wetzstein• 2022

Related benchmarks

TaskDatasetResultRank
Scene ExtrapolationACID (test)
FID52.81
15
Scene ExtrapolationLHQ (test)
FID34.49
6
Novel View SynthesisACID (10 generated sequences)
PSNR23.56
3
Scene ExtrapolationACID
Avg Points Reconstructed3.12e+6
3
Showing 4 of 4 rows

Other info

Code

Follow for update