Cascade-zero123: One image to highly consistent 3d with self-prompted nearby views

Y Chen, J Fang, Y Huang, T Yi, X Zhang, L **e… - … on Computer Vision, 2024 - Springer
Synthesizing multi-view 3D from one single image is a significant but challenging task. Zero-
1-to-3 methods have achieved great success by lifting a 2D latent diffusion model to the 3D …

Dimvis: Diffusion-based multi-view synthesis

G Di Giacomo, G Franzese, T Cerquitelli… - ICML 2024 Workshop …, 2024 - openreview.net
Multi-view observations offer a broader perception of the real world, compared to
observations acquired from a single viewpoint. While existing multi-view 2D diffusion models …

LiftImage3D: Lifting Any Single Image to 3D Gaussians with Video Generation Priors

Y Chen, C Yang, J Fang, X Zhang, L **e… - arxiv preprint arxiv …, 2024 - arxiv.org
Single-image 3D reconstruction remains a fundamental challenge in computer vision due to
inherent geometric ambiguities and limited viewpoint information. Recent advances in …

Dual Dreamer: Extending Single-View Dreamer with Few Shot of Complementary Views

Z Zhang, P Qiao, D Yong, S Liu, W Li, L Cao… - … Conference on Artificial …, 2024 - Springer
Recently, diffusion models demonstrate the potential of generating consistent novel views
from a single view. Due to the limited object information contained in a single input view …