Semi-Supervised Confidence-Level-based Contrastive Discrimination for Class-Imbalanced Semantic Segmentation

Liu, Kangcheng

doi:10.1109/cyber55403.2022.9907146

Cited by 8 publications

(6 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our network framework and innovatively designed modules to merge over-divided clusters to provide pseudo labels are illustrated in Subsection B and C for segmentation and detection respectively. Also, the LiDAR-based approaches are of significance to many industrial applications such as UAV, UGV as well as service robotic navigation as well as inspections [1]- [3] and robotic enhanced large-scale localization in the diverse complex environments [4]- [12], and large-scale robotic semantic scene parsing [13]- [15], as well as robotic control as well as robotic manipulation applications [16]- [21], etc.…”

Section: Proposed Methodologymentioning

confidence: 99%

FG-Net: A Fast and Accurate Framework for Large-Scale LiDAR Point Cloud Understanding

Liu¹,

Gao²,

Feng³

et al. 2023

IEEE Trans. Cybern.

View full text Add to dashboard Cite

Existing state-of-the-art 3D point clouds understanding methods only perform well in a fully supervised manner. To the best of our knowledge, there exists no unified framework which simultaneously solves the downstream highlevel understanding tasks, especially when labels are extremely limited. This work presents a general and simple framework to tackle point clouds understanding when labels are limited. We propose a novel unsupervised region expansion based clustering method for generating clusters. More importantly, we innovatively propose to learn to merge the over-divided clusters based on the local low-level geometric property similarities and the learned high-level feature similarities supervised by weak labels. Hence, the true weak labels guide pseudo labels merging taking both geometric and semantic feature correlations into consideration. Finally, the self-supervised reconstruction and data augmentation optimization modules are proposed to guide the propagation of labels among semantically similar points within a scene. Experimental Results demonstrate that our framework has the best performance among the three most important weakly supervised point clouds understanding tasks including semantic segmentation, instance segmentation, and object detection even when limited points are labeled, under the data-efficient settings for the large-scale 3D semantic scene parsing. The developed techniques have postentials to be applied to downstream tasks for better representations in robotic manipulation and robotic autonomous navigation. Codes and models are publicly available at: https://github.com/KangchengLiu.

show abstract

Section: Proposed Methodologymentioning

confidence: 99%

FG-Net: A Fast and Accurate Framework for Large-Scale LiDAR Point Cloud Understanding

Liu¹,

Gao²,

Feng³

et al. 2023

IEEE Trans. Cybern.

View full text Add to dashboard Cite

show abstract

“…Also, it has recently be studied that the image can bridge the big semantic gap between the modalities of text and 3D shapes. Also, the LiDAR-based approaches are of significance to many industrial applications such as UAV/robotics inspections [31,32,36] and robotic enhanced large-scale localization in the diverse complex environments [26, 28-30, 35, 41, 42, 46, 49], and large-scale robotic scene parsing [37][38][39], as well as robotic control as well as manipulation applications [27,33,34,43,45,50], etc. Differently, we make the first attempt in traditional and learnt 3D descriptor guided weakly supervised point cloud segmentation.…”

Section: Learning-based Point Clouds Understanding Methodsmentioning

confidence: 99%

“…The weakly supervised approaches for point cloud understanding are effective manners to reduce high annotation burdens [32]. Many preliminary attempts have been tried including labeling a small portion of points [16,24,52,73] or semantic classes [69].…”

Section: Weakly Supervised Methods For Point Clouds Understandingmentioning

confidence: 99%

RM3D: Robust Data-Efficient 3D Scene Parsing via Traditional and Learnt 3D Descriptors-Based Semantic Region Merging

Liu

2022

Int J Comput Vis

Self Cite

View full text Add to dashboard Cite

Existing state-of-the-art 3D point cloud understanding methods merely perform well in a fully supervised manner. To the best of our knowledge, there exists no unified framework that simultaneously solves the downstream high-level understanding tasks including both segmentation and detection, especially when labels are extremely limited. This work presents a general and simple framework to tackle point cloud understanding when labels are limited. The first contribution is that we have done extensive methodology comparisons of traditional and learned 3D descriptors for the task of weakly supervised 3D scene understanding, and validated that our adapted traditional PFH-based 3D descriptors show excellent generalization ability across different domains. The second contribution is that we proposed a learning-based region merging strategy based on the affinity provided by both the traditional/learned 3D descriptors and learned semantics. The merging process takes both low-level geometric and high-level semantic feature correlations into consideration. Experimental results demonstrate that our framework has the best performance among the three most important weakly supervised point clouds understanding tasks including semantic segmentation, instance segmentation, and object detection even when very limited number of points are labeled. Our method, termed Region Merging 3D (RM3D), has superior performance on Scan-Net data-efficient learning online benchmarks and other four large-scale 3D understanding benchmarks under various experimental settings, outperforming current arts by a margin for various 3D understanding tasks without complicated learning strategies such as active learning.

show abstract

“…Most importantly, a significant module for Unmanned Aerial Vehicle (UAV) intelligent inspection systems is to develop computer vision algorithms for processing images captured and detecting cracks and structural damages. The visual sensors have been emerging as novel techniques for modeling the environment [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], and has been widely adopted for robotic applications. Various visual sensors and Light Detection And Ranging Sensors (LiDAR) sensors have been widely deployed in various robotics platforms such as automatically navigated aerial robots and ground robots [16], [17], [18], [19], [20], [21].…”

Section: B the Overall System For Uav Inspectionsmentioning

confidence: 99%

“…Various visual sensors and Light Detection And Ranging Sensors (LiDAR) sensors have been widely deployed in various robotics platforms such as automatically navigated aerial robots and ground robots [16], [17], [18], [19], [20], [21]. For example, the SLAM is widely adopted for robot localization, mapping, and navigation applications [11], [22], [23], [12], [24], [25], [26], [27], [28], [29], [30], [31], [22].…”

Section: B the Overall System For Uav Inspectionsmentioning

confidence: 99%

Industrial UAV-Based Unsupervised Domain Adaptive Crack Recognitions: From Database Towards Real-Site Infrastructural Inspections

Liu

Chen

2023

IEEE Trans. Ind. Electron.

View full text Add to dashboard Cite

Automatic crack detection and segmentation play a significant role in the whole system of unmanned aerial vehicle inspections. In this paper, we have implemented a deep learning framework for crack detection based on classical network architectures including Alexnet, VGG, and Resnet. Moreover, inspired by the feature pyramid network architecture, a hierarchical convolutional neural network (CNN) deep learning framework which is efficient in crack segmentation is also proposed, and its performance of it is compared with other state-of-the-art network architecture. We have summarized the existing crack detection and segmentation datasets and established the largest existing benchmark dataset on the internet for crack detection and segmentation, which is open-sourced for the research community. Our feature pyramid crack segmentation network is tested on the benchmark dataset and gives satisfactory segmentation results. A framework for automatic unmanned aerial vehicle inspections is also proposed and will be established for the crack inspection tasks of various concrete structures. All our self-established dataset and codes are opensourced at: https://github.com/KangchengLiu/Crack-Detectionand-Segmentation-Dataset-for-UAV-Inspection.

show abstract

Semi-Supervised Confidence-Level-based Contrastive Discrimination for Class-Imbalanced Semantic Segmentation

Cited by 8 publications

References 21 publications

FG-Net: A Fast and Accurate Framework for Large-Scale LiDAR Point Cloud Understanding

FG-Net: A Fast and Accurate Framework for Large-Scale LiDAR Point Cloud Understanding

RM3D: Robust Data-Efficient 3D Scene Parsing via Traditional and Learnt 3D Descriptors-Based Semantic Region Merging

Industrial UAV-Based Unsupervised Domain Adaptive Crack Recognitions: From Database Towards Real-Site Infrastructural Inspections

Contact Info

Product

Resources

About