High-Resolution Shape Completion Using Deep Neural Networks for Global Structure and Local Geometry Inference

@article{Han2017HighResolutionSC,
  title={High-Resolution Shape Completion Using Deep Neural Networks for Global Structure and Local Geometry Inference},
  author={Xiaoguang Han and Z. Li and Haibin Huang and Evangelos Kalogerakis and Yizhou Yu},
  journal={2017 IEEE International Conference on Computer Vision (ICCV)},
  year={2017},
  pages={85-93}
}
  • Xiaoguang Han, Z. Li, Yizhou Yu
  • Published 22 September 2017
  • Computer Science
  • 2017 IEEE International Conference on Computer Vision (ICCV)
We propose a data-driven method for recovering missing parts of 3D shapes. Our method is based on a new deep learning architecture consisting of two sub-networks: a global structure inference network and a local geometry refinement network. The global structure inference network incorporates a long short-term memorized context fusion module (LSTM-CF) that infers the global structure of the shape based on multi-view depth information provided as part of the input. It also includes a 3D fully… 
A Deep Learning Method Based on Structure Inference for Single-view 3D Reconstruction
  • Wang Ling, Dongzhi He
  • Computer Science
    2021 IEEE 5th Advanced Information Technology, Electronic and Automation Control Conference (IAEAC)
  • 2021
TLDR
A new method 3D-STR is proposed that combines deep learning models and structure inference networks and uses the object structure inferred from the input image to correct the cumulative errors in the reconstruction stage.
Deep Octree-based CNNs with Output-Guided Skip Connections for 3D Shape and Scene Completion
TLDR
A novel output-guided skip-connection is introduced to the network structure for better preserving the input geometry and learning geometry prior from data effectively and this network achieves state-of-the-art results in 3D shape completion and semantic scene computation.
STD-Net: Structure-preserving and Topology-adaptive Deformation Network for 3D Reconstruction from a Single Image
TLDR
Experimental results on the images from ShapeNet show that the proposed STD-Net has better performance than other state-of-the-art methods onconstructing 3D objects with complex structures and fine geometric details.
Unsupervised Shape Completion via Deep Prior in the Neural Tangent Kernel Perspective
TLDR
This article interprets the deep prior from a neural tangent kernel (NTK) perspective and shows that the completed shape patches by the trained CNN are naturally similar to existing patches, as they are proximate in the kernel feature space induced by NTK.
High-Quality Textured 3D Shape Reconstruction with Cascaded Fully Convolutional Networks
TLDR
Qualitative and quantitative experimental results on both synthetic and real-world datasets demonstrate that the presented approach outperforms existing state-of-the-art work regarding visual quality and accuracy of reconstructed models.
Denoise and Contrast for Category Agnostic Shape Completion
TLDR
A deep learning model that exploits the power of self-supervision to perform 3D point cloud completion, estimating the missing part and a context region around it and decoding the obtained joint representation confirms how this approach is able to work on novel categories without relying neither on classification and shape symmetry priors, nor on adversarial training procedures.
ASHF-Net: Adaptive Sampling and Hierarchical Folding Network for Robust Point Cloud Completion
TLDR
This work proposes a denoising auto-encoder with an adaptive sampling module and a hierarchical folding decoder with the gated skip-attention and multi-resolution completion goal to effectively exploit the local structure details of partial inputs.
Cascaded Refinement Network for Point Cloud Completion with Self-supervision.
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.
ForkNet: Multi-Branch Volumetric Semantic Completion From a Single Depth Image
We propose a novel model for 3D semantic completion from a single depth image, based on a single encoder and three separate generators used to reconstruct different geometric and semantic
Learning to Reconstruct High-Quality 3D Shapes with Cascaded Fully Convolutional Networks
TLDR
A novel cascaded 3D convolutional network architecture is introduced, which learns to reconstruct implicit surface representations from noisy and incomplete depth maps in a progressive, coarse-to-fine manner.
...
...

References

SHOWING 1-10 OF 34 REFERENCES
Shape Completion Using 3D-Encoder-Predictor CNNs and Shape Synthesis
TLDR
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.
3D ShapeNets: A deep representation for volumetric shapes
TLDR
This work proposes to represent a geometric 3D shape as a probability distribution of binary variables on a 3D voxel grid, using a Convolutional Deep Belief Network, and shows that this 3D deep representation enables significant performance improvement over the-state-of-the-arts in a variety of tasks.
Data-driven structural priors for shape completion
TLDR
This work focuses on reconstructing complete geometry from a single scan acquired with a low-quality consumer-level scanning device, using a collection of example 3D shapes to build structural part-based priors that are necessary to complete the shape.
A Field Model for Repairing 3D Shapes
TLDR
Experimental results have shown the robustness and efficiency of the proposed method in repairing noisy and incomplete 3D shapes.
LSTM-CF: Unifying Context Modeling and Fusion with LSTMs for RGB-D Scene Labeling
TLDR
This paper addresses the problem of semantic labeling of RGB-D scenes by developing a novel Long Short-Term Memorized Context Fusion (LSTM-CF) Model that captures and fuses contextual information from multiple channels of photometric and depth data, and incorporating this model into deep convolutional neural networks (CNNs) for end-to-end training.
Multimodal deep learning for robust RGB-D object recognition
TLDR
This paper leverages recent progress on Convolutional Neural Networks (CNNs) and proposes a novel RGB-D architecture for object recognition that is composed of two separate CNN processing streams - one for each modality - which are consecutively combined with a late fusion network.
ShapeNet: An Information-Rich 3D Model Repository
TLDR
ShapeNet contains 3D models from a multitude of semantic categories and organizes them under the WordNet taxonomy, a collection of datasets providing many semantic annotations for each 3D model such as consistent rigid alignments, parts and bilateral symmetry planes, physical sizes, keywords, as well as other planned annotations.
GlobFit: consistently fitting primitives by discovering global relations
TLDR
This work introduces an algorithm to directly couple the local and global aspects of the problem, and introduces a method that simultaneously recovers a set of locally fitted primitives along with their global mutual relations.
Structure recovery by part assembly
This paper presents a technique that allows quick conversion of acquired low-quality data from consumer-level scanning devices to high-quality 3D models with labeled semantic parts and meanwhile
Example-based 3D scan completion
TLDR
This work retrieves suitable context models from the database, warps the retrieved models to conform with the input data, and consistently blends the warped models to obtain the final consolidated 3D shape.
...
...