74 citations · 209 across the 15 of their papers we have counts for
15 papers
Retro-FPN: Retrospective Feature Pyramid Network for Point Cloud Semantic Segmentation
Peng Xiang, Xin Wen, Yu-Shen Liu +3
Learning per-point semantic features from the hierarchical feature pyramid is essential for point cloud semantic segmentation. However, most previous methods suffered from ambiguou…
Learning Deep Implicit Functions for 3D Shapes with Dynamic Code Clouds
Tianyang Li, Xin Wen, Yu-Shen Liu +2
Deep Implicit Function (DIF) has gained popularity as an efficient 3D shape representation. To capture geometry details, current methods usually learn DIF using local latent codes,…
3D Shape Reconstruction from 2D Images with Disentangled Attribute Flow
Xin Wen, Junsheng Zhou, Yu-Shen Liu +2
Reconstructing 3D shape from a single 2D image is a challenging task, which needs to estimate the detailed 3D structures based on the semantic attributes from 2D image. So far, mos…
3D-OAE: Occlusion Auto-Encoders for Self-Supervised Learning on Point Clouds
Junsheng Zhou, Xin Wen, Baorui Ma +4
The manual annotation for large-scale point clouds is still tedious and unavailable for many harsh real-world tasks. Self-supervised learning, which is used on raw and unlabeled da…
PMP-Net++: Point Cloud Completion by Transformer-Enhanced Multi-step Point Moving Paths
Xin Wen, Peng Xiang, Zhizhong Han +4
Point cloud completion concerns to predict missing part for incomplete 3D shapes. A common strategy is to generate complete shape according to incomplete input. However, unordered…
Snowflake Point Deconvolution for Point Cloud Completion and Generation with Skip-Transformer
Peng Xiang, Xin Wen, Yu-Shen Liu +4
Most existing point cloud completion methods suffer from the discrete nature of point clouds and the unstructured prediction of points in local regions, which makes it difficult to…