The need to model visual information with compact representations has existed since the early days of computer vision. We implemented in the past a segmentation and model recovery method for range ...

UniPre3D is the first unified pre-training method for 3D point clouds that effectively handles both object- and scene-level data through cross-modal Gaussian splatting. Our proposed pre-training task ...

This work addresses the problem of 3D human pose and shape estimation from a sequence of point clouds. Existing sequential 3D human shape estimation methods mainly focus on the template model fitting ...

PointTransformer-SimCLR for 3D Point Clouds We train a SimCLR-style contrastive model for 3D point clouds, using a Point Transformer–based encoder to capture both local and global geometry. This ...

In this work, we propose the Multi-level Graph Convolutional Neural Network (MLGCN), an ultra-efficient model for 3D point cloud analysis. The MLGCN model utilizes shallow Graph Neural Network (GNN) ...

To simplify these processes, technological advancements have provided a solution: site surveys based on point clouds and 3D scanning, which have the potential to revolutionize the design process.

Point clouds have evolved into one of the most important data formats for 3D representation. It is becoming more popular as a result of the increasing affordability of acquisition equipment and ...

SDS-Complete leverages a pre-trained text-to-image diffusion model to guide the completion of missing parts in point clouds. Traditional approaches to point cloud completion rely heavily on ...