SegGroup: Seg-Level Supervision for 3D Instance and Semantic Segmentation

  title={SegGroup: Seg-Level Supervision for 3D Instance and Semantic Segmentation},
  author={An Tao and Yueqi Duan and Yi Wei and Jiwen Lu and Jie Zhou},
  journal={IEEE transactions on image processing : a publication of the IEEE Signal Processing Society},
  • An Tao, Yueqi Duan, Jie Zhou
  • Published 18 December 2020
  • Computer Science
  • IEEE transactions on image processing : a publication of the IEEE Signal Processing Society
Most existing point cloud instance and semantic segmentation methods rely heavily on strong supervision signals, which require point-level labels for every point in the scene. However, such strong supervision suffers from large annotation costs, arousing the need to study efficient annotating. In this paper, we discover that the locations of instances matter for both instance and semantic 3D scene segmentation. By fully taking advantage of locations, we design a weakly supervised point cloud… 
Dual Adaptive Transformations for Weakly Supervised Point Cloud Segmentation
This paper proposes a novel DAT model for weakly supervised point cloud segmentation, where the dual adaptive transformations are performed via an adversarial strategy at both point-level and region-level, aiming at enforcing the local and structural smoothness constraints on 3D point clouds.


One Thing One Click: A Self-Training Approach for Weakly Supervised 3D Semantic Segmentation
This work designs a novel self-training approach, in which the annotator only needs to label one point per object, and outperforms all existing weakly supervised methods for 3D semantic segmentation by a large margin, and is comparable to those of the fully supervised counterparts.
Exploring Data-Efficient 3D Scene Understanding with Contrastive Scene Contexts
This study reveals that exhaustive labelling of 3D point clouds might be unnecessary; and remarkably, on ScanNet, even using 0.1% of point labels, the method achieves state-of-the-art results on a suite of benchmarks where training data or labels are scarce.
PointGroup: Dual-Set Point Grouping for 3D Instance Segmentation
This paper designs a two-branch network to extract point features and predict semantic labels and offsets, for shifting each point towards its respective instance centroid, and presents PointGroup, a new end-to-end bottom-up architecture specifically focused on better grouping the points by exploring the void space between objects.
Multi-Path Region Mining for Weakly Supervised 3D Semantic Segmentation on Point Clouds
This paper introduces a multi-path region mining module to generate pseudo point-level labels from a classification network trained with weak labels, and uses the point- level pseudo label to train a point cloud segmentation network in a fully supervised manner.
Dynamic Graph CNN for Learning on Point Clouds
This work proposes a new neural network module suitable for CNN-based high-level tasks on point clouds, including classification and segmentation called EdgeConv, which acts on graphs dynamically computed in each layer of the network.
Semi-Supervised Classification with Graph Convolutional Networks
A scalable approach for semi-supervised learning on graph-structured data that is based on an efficient variant of convolutional neural networks which operate directly on graphs which outperforms related methods by a significant margin.
Large-Scale Machine Learning with Stochastic Gradient Descent
A more precise analysis uncovers qualitatively different tradeoffs for the case of small-scale and large-scale learning problems.
Perturbed Self-Distillation: Weakly Supervised Large-Scale Point Cloud Semantic Segmentation
This work proposes a perturbed self-distillation (PSD) framework, inspired by self-supervised learning, and presents a well-integrated context-aware module to explicitly regularize the affinity correlation of labeled points.
Mix3D: Out-of-Context Data Augmentation for 3D Scenes
This work proposes a “mixing” technique which creates new training samples by combining two augmented scenes, making it harder for models to rely on scene context alone, and instead infer semantics from local structure as well.
Instance Segmentation in 3D Scenes using Semantic Superpoint Tree Networks
An end-to-end solution of Semantic Superpoint Tree Network (SSTNet) for proposing object instances from scene points, and design in SSTNet a refinement module, termed CliqueNet, to prune superpoints that may be wrongly grouped into instance proposals.