Cascaded Refinement Network for Point Cloud Completion With Self-Supervision

  title={Cascaded Refinement Network for Point Cloud Completion With Self-Supervision},
  author={Xiaogang Wang and Marcelo H. Ang and Gim Hee Lee},
  journal={IEEE Transactions on Pattern Analysis and Machine Intelligence},
  • Xiaogang WangM. AngG. Lee
  • Published 17 October 2020
  • Computer Science
  • IEEE Transactions on Pattern Analysis and Machine Intelligence
Point clouds are often sparse and incomplete, which imposes difficulties for real-world applications. Existing shape completion methods tend to generate rough shapes without fine-grained details. Considering this, we introduce a two-branch network for shape completion. The first branch is a cascaded shape completion sub-network to synthesize complete objects, where we propose to use the partial input together with the coarse output to preserve the object details during the dense point… 

Cross-modal Learning for Image-Guided Point Cloud Shape Completion

It is shown how it is possible to effectively combine the information from the two modalities in a localized latent space, thus avoiding the need for complex point cloud reconstruction methods from single views used by the state-of-the-art.

Contrastive Learning for 3D Point Clouds Classification and Shape Completion

The idea is to add contrastive learning into AutoEncoders to encourage global feature learning of the point cloud classes by optimizing triplet loss and embeddings generated from the contrastive AutoEncoder enhances shape completion and classification performance.

A Geometry-Enhanced 6D Pose Estimation Network With Incomplete Shape Recovery for Industrial Parts

A geometry-enhanced network with incomplete shape recovery (GER-Net) to estimate the 6D pose of industrial parts and achieves the best performance with remarkable accuracy and robustness.

Snowflake Point Deconvolution for Point Cloud Completion and Generation with Skip-Transformer

  • Peng XiangXin Wen Zhizhong Han
  • Environmental Science, Computer Science
    IEEE transactions on pattern analysis and machine intelligence
  • 2022
The insight into the detailed geometry is to introduce a skip-transformer in the SPD to learn the point splitting patterns that can best fit the local regions, which enables us to predict highly detailed geometries.

Comprehensive Review of Deep Learning-Based 3D Point Cloud Completion Processing and Analysis

This work aims to conduct a comprehensive survey on various methods of point cloud completion, including point-based, view- based, convolution-based), convolution, graph, graph based, generative model based, transformer-based approaches, etc, and summarizes the comparisons among these methods to provoke further research insights.

Implicit and Efficient Point Cloud Completion for 3D Single Object Tracking

This work designs a strong pipeline to extract discriminative features and conducts the matching with the attention mechanism, and introduces two novel modules, i.e., Adaptive Refine Prediction (ARP) and Target Knowledge Transfer (TKT), to tackle the misalignment issue and calls the overall framework PCET.

CP3: Unifying Point Cloud Completion by Pretrain-Prompt-Predict Paradigm

Inspired by prompting approaches from NLP, the CP3 is creatively reinterpreted as the prompting and predicting stages, respectively, and a concise self-supervised pretraining stage before prompting is introduced that can effectively increase robustness of point cloud generation.

VQ-DcTr: Vector-Quantized Autoencoder With Dual-channel Transformer Points Splitting for 3D Point Cloud Completion

Existing point cloud completion methods mainly utilize the global shape representation to recover the missing regions of the 3D shape from the partial point cloud. However, these methods learn the

SCARP: 3D Shape Completion in ARbitrary Poses for Improved Grasping

SCARP is a model that performs canonicalization, pose estimation, and shape completion in a single network, improving the performance by 45% over the existing baselines and is used for improving grasp proposals on tabletop objects.

Lidar Upsampling With Sliced Wasserstein Distance

Lidar became an important component of the perception systems in autonomous driving. But challenges of training data acquisition and annotation made emphasized the role of the sensor to sensor domain



Cascaded Refinement Network for Point Cloud Completion

This work proposes a cascaded refinement network together with a coarse-to-fine strategy to synthesize the detailed object shapes and designs a patch discriminator that guarantees every local area has the same pattern with the ground truth to learn the complicated point distribution.

Detail Preserved Point Cloud Completion via Separated Feature Aggregation

Qualitative and quantitative evaluations demonstrate that the proposed network outperforms current state-of-the art methods especially on detail preservation.

Learning 3D Shape Completion from Laser Scan Data with Weak Supervision

This work proposes a weakly-supervised learning-based approach to 3D shape completion which neither requires slow optimization nor direct supervision and is able to compete with a fully supervised baseline and a state-of-the-art data-driven approach while being significantly faster.

Point Cloud Completion by Skip-Attention Network With Hierarchical Folding

This work proposes Skip-Attention Network (SA-Net), a skip-attention mechanism to effectively exploit the local structure details of incomplete point clouds during the inference of missing parts, and proposes a novel structure-preserving decoder with hierarchical folding for complete shape generation.

SG-NN: Sparse Generative Neural Networks for Self-Supervised Scene Completion of RGB-D Scans

A novel approach that converts partial and noisy RGB-D scans into high-quality 3D scene reconstructions by inferring unobserved scene geometry and combined with a new 3D sparse generative convolutional neural network architecture is able to predict highly detailed surfaces in a coarse-to-fine hierarchical fashion.

PointDAN: A Multi-Scale 3D Domain Adaption Network for Point Cloud Representation

A novel 3D Domain Adaptation Network for point cloud data (PointDAN) is proposed, which jointly aligns the global and local features in multi-level and demonstrates the superiority of the model over the state-of-the-art general-purpose DA methods.

Deformable Shape Completion with Graph Convolutional Autoencoders

This work proposes a novel learning-based method for the completion of partial shapes using a variational autoencoder with graph convolutional operations that learns a latent space for complete realistic shapes that best fits the generated shape to the known partial input.

PCN: Point Completion Network

The experiments show that PCN produces dense, complete point clouds with realistic structures in the missing regions on inputs with various levels of incompleteness and noise, including cars from LiDAR scans in the KITTI dataset.

GRNet: Gridding Residual Network for Dense Point Cloud Completion

This work devise two novel differentiable layers, named Gridding and Gridding Reverse, to convert between point clouds and 3D grids without losing structural information, and presents the differentiable Cubic Feature Sampling layer to extract features of neighboring points, which preserves context information.

Shape Completion Using 3D-Encoder-Predictor CNNs and Shape Synthesis

A data-driven approach to complete partial 3D shapes through a combination of volumetric deep neural networks and 3D shape synthesis and a 3D-Encoder-Predictor Network (3D-EPN) which is composed of 3D convolutional layers.