RefRec: Pseudo-labels Refinement via Shape Reconstruction for Unsupervised 3D Domain Adaptation

  title={RefRec: Pseudo-labels Refinement via Shape Reconstruction for Unsupervised 3D Domain Adaptation},
  author={Adriano Cardace and Riccardo Spezialetti and Pierluigi Zama Ramirez and Samuele Salti and Luigi di Stefano},
  journal={2021 International Conference on 3D Vision (3DV)},
Unsupervised Domain Adaptation (UDA) for point cloud classification is an emerging research problem with relevant practical motivations. Reliance on multi-task learning to align features across domains has been the standard way to tackle it. In this paper, we take a different path and propose RefRec, the first approach to investigate pseudo-labels and self-training in UDA for point clouds. We present two main innovations to make self-training effective on 3D data: i) refinement of noisy pseudo… 


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.
Joint Supervised and Self-Supervised Learning for 3D Real World Challenges
This work proposes to enrich standard feature representations by leveraging self-supervision through a multi-task model that can solve a 3D puzzle while learning the main task of shape classification or part segmentation in synthetic and real world point clouds.
3DMatch: Learning Local Geometric Descriptors from RGB-D Reconstructions
3DMatch is presented, a data-driven model that learns a local volumetric patch descriptor for establishing correspondences between partial 3D data that consistently outperforms other state-of-the-art approaches by a significant margin.
Unsupervised Domain Adaptation for Semantic Segmentation via Class-Balanced Self-training
This paper proposes a novel UDA framework based on an iterative self-training (ST) procedure, where the problem is formulated as latent variable loss minimization, and can be solved by alternatively generating pseudo labels on target data and re-training the model with these labels.
PPF-FoldNet: Unsupervised Learning of Rotation Invariant 3D Local Descriptors
It is demonstrated that despite having six degree-of-freedom invariance and lack of training labels, PPF-FoldNet achieves state of the art results in standard benchmark datasets and outperforms its competitors when rotations and varying point densities are present.
Learning to Adapt Structured Output Space for Semantic Segmentation
A multi-level adversarial network is constructed to effectively perform output space domain adaptation at different feature levels and it is shown that the proposed method performs favorably against the state-of-the-art methods in terms of accuracy and visual quality.
PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation
This paper designs a novel type of neural network that directly consumes point clouds, which well respects the permutation invariance of points in the input and provides a unified architecture for applications ranging from object classification, part segmentation, to scene semantic parsing.
FoldingNet: Point Cloud Auto-Encoder via Deep Grid Deformation
A novel end-to-end deep auto-encoder is proposed to address unsupervised learning challenges on point clouds, and is shown, in theory, to be a generic architecture that is able to reconstruct an arbitrary point cloud from a 2D grid.
Domain Adaptation for Semantic Segmentation With Maximum Squares Loss
The maximum squares loss prevents the training process being dominated by easy-to-transfer samples in the target domain, and introduces the image-wise weighting ratio to alleviate the class imbalance in the unlabeled target domain.
Domain Adaptive Faster R-CNN for Object Detection in the Wild
This work builds on the recent state-of-the-art Faster R-CNN model, and design two domain adaptation components, on image level and instance level, to reduce the domain discrepancy, based on $$-divergence theory.