Get 3D object point cloud from images using diffusion model based on ViT with DiffPoint
Get 3D object point cloud from images using diffusion model based on ViT with DiffPoint
DiffPoint: Single and Multi-view Point Cloud Reconstruction with ViT Based Diffusion Model
arXiv paper abstract https://arxiv.org/abs/2402.11241
arXiv PDF paper https://arxiv.org/pdf/2402.11241.pdf
As … 2D-to-3D reconstruction has gained … attention … it becomes crucial … to generate high-quality point clouds.
… propose … DiffPoint that combines ViT and diffusion models for the task of point cloud reconstruction.
At each diffusion step, … divide the noisy point clouds into irregular patches.
… using a standard ViT backbone that treats all inputs as tokens (including time information, image embeddings, and noisy patches), … train … model to predict target points based on input images.
… evaluate DiffPoint on both single-view and multi-view reconstruction tasks and achieve state-of-the-art results.
… introduce a unified and flexible feature fusion module for aggregating image features from single or multiple input images …
Stay up to date. Subscribe to my posts https://morrislee1234.wixsite.com/website/contact
Web site with my other posts by category https://morrislee1234.wixsite.com/website
LinkedIn https://www.linkedin.com/in/morris-lee-47877b7b
