ABSTRACT
The task of 3D point cloud completion is to predict a complete point cloud from the incomplete partial point cloud. Generally, the encoder is used to extract the global shape features of the input incomplete point cloud, and then the decoder infers the complete point cloud. At present, some methods have been improved by multi-resolution encoders and multi-layer decoders, and achieved obvious results. However, these methods still cannot fully express the shape features. In order to solve this problem, we propose a feature fusion mechanism based on skip connection. The features extracted from each resolution point cloud are connected with the input of corresponding decoder. Then they are weighted and fused to obtain denser features, which can be decoded into a finer point cloud. In addition, the current loss function is still not a good measure of the similarity between two point clouds, so we also proposed a multi-stage local average Hausdroff Loss to form a joint reconstruction loss function to guide the generation of missing point clouds. Experimental results prove the effectiveness of our method in point cloud completion tasks, and show that it products better performance than existing methods.
- He K, Zhang X, Ren S, et al. 2016. Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition. 770--778.Google Scholar
- Ren S, He K, Girshick R, et al. 2015. Faster r-cnn: Towards real-time object detection with region proposal networks. In Advances in neural information processing systems. 91--99.Google Scholar
- de Vos B D, Berendsen F F, Viergever M A, et al. 2017. End-to-end unsupervised deformable image registration with a convolutional neural network. Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support. Springer, Cham, 204--212.Google Scholar
- He K, Gkioxari G, Dollár P, et al. 2017. Mask r-cnn. In Proceedings of the IEEE international conference on computer vision. 2961--2969.Google Scholar
- Lin T Y, Dollár P, Girshick R, et al. 2017. Feature pyramid networks for object detection. In Proceedings of the IEEE conference on computer vision and pattern recognition. 2117--2125.Google Scholar
- Wang X, Guo Q, Zhao X. 2020. Multiple Spaces Deep Hashing for Image Retrieval. In 2020 12th International Conference on Advanced Computational Intelligence (ICACI). IEEE, 397--401.Google Scholar
- Deng J, Dong W, Socher R, et al. 2009. Imagenet: A large-scale hierarchical image database. In 2009 IEEE conference on computer vision and pattern recognition. Ieee, 248--255.Google Scholar
- Qi C R, Su H, Mo K, et al. 2017. Pointnet: Deep learning on point sets for 3d classification and segmentation. In Proceedings of the IEEE conference on computer vision and pattern recognition. 652--660.Google Scholar
- Achlioptas P, Diamanti O, Mitliagkas I, et al. 2018. Learning representations and generative models for 3d point clouds. In International conference on machine learning. PMLR, 40--49.Google Scholar
- Yuan W, Khot T, Held D, et al. 2018. Pcn: Point completion network. In 2018 International Conference on 3D Vision (3DV). IEEE, 728--737.Google Scholar
- Sarmad M, Lee H J, Kim Y M. 2019. Rl-gan-net: A reinforcement learning agent controlled gan network for real-time point cloud shape completion. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 5898--5907.Google ScholarCross Ref
- Yang Y, Feng C, Shen Y, et al. 2018. Foldingnet: Point cloud auto-encoder via deep grid deformation. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 206--215.Google Scholar
- Zhao Y, Birdal T, Deng H, et al. 2019. 3D point capsule networks. In Proceedings of the IEEE conference on computer vision and pattern recognition. 1009--1018.Google Scholar
- Sung M, Kim V G, Angst R, et al. 2015. Data-driven structural priors for shape completion. ACM Transactions on Graphics (TOG), 34(6): 1--11.Google ScholarDigital Library
- Thanh Nguyen D, Hua B S, Tran K, et al. 2016. A field model for repairing 3d shapes. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 5676--5684.Google Scholar
- Kalogerakis E, Chaudhuri S, Koller D, et al. 2012. A probabilistic model for component-based shape synthesis. ACM Transactions on Graphics (TOG), 31(4): 1--11.Google ScholarDigital Library
- Dai A, Ruizhongtai Qi C, Nießner M. 2017. Shape completion using 3d-encoder-predictor cnns and shape synthesis. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 5868--5877.Google ScholarCross Ref
- Su H, Maji S, Kalogerakis E, et al. 2015. Multi-view convolutional neural networks for 3d shape recognition. In Proceedings of the IEEE international conference on computer vision. 945--953.Google Scholar
- Stutz D, Geiger A. 2018. Learning 3d shape completion from laser scan data with weak supervision. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 1955--1964.Google ScholarCross Ref
- Qi C R, Yi L, Su H, et al. 2017. Pointnet++: Deep hierarchical feature learning on point sets in a metric space. Advances in neural information processing systems. 5099--5108.Google Scholar
- Yi L, Kim V G, Ceylan D, et al. 2016. A scalable active framework for region annotation in 3d shape collections. ACM Transactions on Graphics (ToG), 35(6): 1--12.Google ScholarDigital Library
- Gadelha M, Wang R, Maji S. 2018. Multiresolution tree networks for 3d point cloud processing. Proceedings of the European Conference on Computer Vision (ECCV). 103--118.Google ScholarDigital Library
- Huang Z, Yu Y, Xu J, et al. 2020. PF-Net: Point Fractal Network for 3D Point Cloud Completion. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 7662--7670.Google Scholar
Index Terms
- Multi-resolution Dense Network for Point Cloud Completion
Recommendations
High-fidelity point cloud completion with low-resolution recovery and noise-aware upsampling
AbstractCompleting an unordered partial point cloud is a challenging task. Existing approaches that rely on decoding a latent feature to recover the complete shape, often lead to the completed point cloud being over-smoothing, losing details, ...
Graphical abstractDisplay Omitted
MRAC-Net: Multi-resolution Anisotropic Convolutional Network for 3D Point Cloud Completion
PRICAI 2021: Trends in Artificial IntelligenceAbstractPoint cloud completion aims to infer the missing parts of the 3D object from incomplete point clouds. Previous methods usually use Multi-layer Perceptrons to directly extract latent features from incomplete point clouds. However, these latent ...
TNT-Net: Point Cloud Completion by Transformer in Transformer
MultiMedia ModelingAbstractEstimating the overall structure of a point cloud from a partial 3D point cloud input is a crucial task in computer vision. However, existing point cloud completion methods often overlook object detail information and the local correlation within ...
Comments