Abstract:There has been a growing interest in the research of semantic labeling on scenes represented by 3D point clouds. A fundamental issue that has been largely ignored is the unavoidable presence of unknown objects and the lack of effective ways of dealing with them. Traditional methods usually label unknown objects as one of the pre-trained classes which is either a meaningful target class or a defined unknown class that collectively refers to all uninterested objects. Due to the fact that the class of unknown in essence is a collection of many unseen or uninterested classes, in which the in-class variation is significant and less manageable. It is challenging to solve the unknown problem in a pre-trained manner. In order to advance the research on semantic labeling with the presence of unknown objects, this study investigates the feasibility of adopting an open-set approach, i.e. train a model without unknown objects and reject them accurately in the test. In this paper, we propose a method that exploits the conflict of different labeling results in order to withstand the negative effect of unknown objects. The proposed framework relies on a Conditional Random Field (CRF) to capture inherent spatial relationships and appearance similarities between objects, and employs a Probability of Inclusion Support Vector Machine (PISVM) to estimate an unknown likelihood for each training class. The probabilistic outputs from both CRF and PISVM are then proposed to be combined under the Dempster Shafer theory for conflict measurement and unknown rejection. The novelty lies in that the method encodes both contextual constrains and unknown likelihood for performance enhancement. Comprehensive experimental results on publicly available data sets are presented to show the negative effects of unknown objects and the improvements on labeling accuracy achieved by the proposed method.

Hybrid Clustering of Data and Vague Concepts Based on Labels Semantics

Linguistic Modelling Based on Semantic Similarity Relation among Linguistic Labels

Uncertainty Modeling For Data Mining A Label Semantics Approach Introduction

Learning Fuzzy Semantic Cell by Principles of Maximum Coverage, Maximum Specificity, and Maximum Fuzzy Entropy of Vague Concept

Towards Open-Set Semantic Labeling in 3D Point Clouds : Analysis on the Unknown Class.

Learning Disjunctive Concepts Based on Fuzzy Semantic Cell Models Through Principles of Justifiable Granularity and Maximum Fuzzy Entropy

Labeling Clusters from Both Linguistic and Statistical Perspectives: A Hybrid Approach

The interelation in fuzzy clustering

Fuzzy-Based Concept Learning Method: Exploiting Data With Fuzzy Conceptual Clustering

Clustering by Heterogeneous Data Fusion : Framework and Applications

PointCluster: Deep Clustering of 3-D Point Clouds With Semantic Pseudo-Labeling

Image Annotations Based on Semi-supervised Clustering with Semantic Soft Constraints.

Data Clustering using a Hybrid of Fuzzy C-Means and Quantum-behaved Particle Swarm Optimization

Augmentation of Soft Partition with a Granular Prototype Based Fuzzy C-Means

Compositional Clustering: Applications to Multi-Label Object Recognition and Speaker Identification

A Group-Based Distance Learning Method for Semisupervised Fuzzy Clustering

Semantic-Enhanced Image Clustering

Validating Clustering for Gene Expression Data Based on Semantic Distance of Gene Ontology Terms

On Conceptual Labeling of a Bag of Words.

Fuzzy c-Means Clustering with Discriminative Projection

An Optimization Model for Fuzzy Binary Clustering