The reviewed record of science sign in
Pith

arxiv: 2309.11525 · v3 · pith:DEXDUB35 · submitted 2023-09-20 · cs.CV

Light Field Diffusion for Single-View Novel View Synthesis

Reviewed by Pith T0 review T1 audit T2 compute T3 formal T4 kernel pith:DEXDUB35record.jsonopen to challenge →

classification cs.CV
keywords diffusionimagesfieldimagelightnovelviewcamera
0
0 comments X
read the original abstract

Single-view novel view synthesis (NVS), the task of generating images from new viewpoints based on a single reference image, is important but challenging in computer vision. Recent advancements in NVS have leveraged Denoising Diffusion Probabilistic Models (DDPMs) for their exceptional ability to produce high-fidelity images. However, current diffusion-based methods typically utilize camera pose matrices to globally and implicitly enforce 3D constraints, which can lead to inconsistencies in images generated from varying viewpoints, particularly in regions with complex textures and structures. To address these limitations, we present Light Field Diffusion (LFD), a novel conditional diffusion-based approach that transcends the conventional reliance on camera pose matrices. Starting from the camera pose matrices, LFD transforms them into light field encoding, with the same shape as the reference image, to describe the direction of each ray. By integrating light field encoding with the reference image, our method imposes local pixel-wise constraints within the diffusion process, fostering enhanced view consistency. Our approach not only involves training image LFD on the ShapeNet Car dataset but also includes fine-tuning a pre-trained latent diffusion model on the Objaverse dataset. This enables our latent LFD model to exhibit remarkable zero-shot generalization capabilities across out-of-distribution datasets like RTMV as well as in-the-wild images. Experiments demonstrate that LFD not only produces high-fidelity images but also achieves superior 3D consistency in complex regions, outperforming existing novel view synthesis methods.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.