ReDAL: Region-based and Diversity-aware Active Learning for Point Cloud Semantic Segmentation

Liu, Yueh-Cheng; Huang, Yu‐Kai; Lee, Hsin-Ying; Su, Hung-Ting; Huang, Ping-Chia; Hsu, Winston H.

doi:10.1109/iccv48922.2021.01522

Cited by 55 publications

(40 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Table 1 reports that SSDR-AL reduces the annotation cost by up to 63.0% lower compared to the baseline ClassBal [3] in achieving the 90% performance of fully supervised learning. Besides, SSDR-AL can annotate 1.3% fewer points than ReDAL [30] in achieving the 90%performance of fully supervised learning reported in Table 2. Semantic3D.…”

Section: Comparison With State-of-the-art Methodsmentioning

confidence: 95%

“…ReDAL [30] SPVCNN [27] 13% ReDAL [30] MinkowskiNet [5] 15% Random Randlanet [11] 40.9% Entropy [13] Randlanet [11] 46.7% BvSB [13] Randlanet [11] 43.0% ClassBal [3] Randlanet [11] 13.3% SSDR-AL(Ours)…”

Section: Methodsmentioning

confidence: 99%

“…The labeled superpoints and sub-regions will be added into the labeled set, and all of labeled or partially labeled superpoints will be deleted from the unlabeled set. Meanwhile, the other superpoints will be returned to the un- indicates the results are reported in original paper [30].…”

Section: Implementation Detailsmentioning

confidence: 99%

See 2 more Smart Citations

Active Learning for Point Cloud Semantic Segmentation via Spatial-Structural Diversity Reasoning

Shao¹,

Luo²,

Liu³

et al. 2022

Preprint

View full text Add to dashboard Cite

The expensive annotation cost is notoriously known as a main constraint for the development of the point cloud semantic segmentation technique. In this paper, we propose a novel active learning-based method to tackle this problem. Dubbed SSDR-AL, our method groups the original point clouds into superpoints and selects the most informative and representative ones for label acquisition. We achieve the selection mechanism via a graph reasoning network that considers both the spatial and structural diversity of the superpoints. To deploy SSDR-AL in a more practical scenario, we design a noise aware iterative labeling scheme to confront the "noisy annotation" problem introduced by previous dominant labeling methods in superpoints. Extensive experiments on two point cloud benchmarks demonstrate the effectiveness of SSDR-AL in the semantic segmentation task. Particularly, SSDR-AL significantly outperforms the baseline method when the labeled sets are small, where SSDR-AL requires only 5.7% and 1.9% annotation costs to achieve the performance of 90% fully supervised learning on S3DIS and Semantic3D datasets, respectively.

show abstract

Section: Comparison With State-of-the-art Methodsmentioning

confidence: 95%

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Active Learning for Point Cloud Semantic Segmentation via Spatial-Structural Diversity Reasoning

Shao¹,

Luo²,

Liu³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…However, very few AL studies are conducted for point cloud semantic segmentation. Authors (Wu et al, 2021) evaluate uncertainty and diversity-based approaches for point cloud semantic segmentation. This study is the closest to our current work.…”

Section: Related Workmentioning

confidence: 99%

LiDAR Dataset Distillation within Bayesian Active Learning Framework Understanding the Effect of Data Augmentation

Duong¹,

Almin²,

Lemarié³

et al. 2022

Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Application

View full text Add to dashboard Cite

Autonomous driving (AD) datasets have progressively grown in size in the past few years to enable better deep representation learning. Active learning (AL) has re-gained attention recently to address reduction of annotation costs and dataset size. AL has remained relatively unexplored for AD datasets, especially on point cloud data from LiDARs. This paper performs a principled evaluation of AL based dataset distillation on (1/4th) of the large Semantic-KITTI dataset. Further on, the gains in model performance due to data augmentation (DA) are demonstrated across different subsets of the AL loop. We also demonstrate how DA improves the selection of informative samples to annotate. We observe that data augmentation achieves full dataset accuracy using only 60% of samples from the selected dataset configuration. This provides faster training time and subsequent gains in annotation costs.

show abstract

“…Common strategies include uncertainty sampling [14,18,51] and representative sampling [1,49,57]. While label acquisition for dense prediction tasks such as segmentation is more expensive and laborious than image classification, there has been considerably less work [2,3,21,52,56,75]. A recent example in [3] proposes to adaptively select image regions based on reinforcement learning, which is a more efficient way than labeling entire images.…”

Section: Related Workmentioning

confidence: 99%

Towards Fewer Annotations: Active Learning via Region Impurity and Prediction Uncertainty for Domain Adaptive Semantic Segmentation

Xie¹,

Yuan²,

Li³

et al. 2021

Preprint

View full text Add to dashboard Cite

Self-training has greatly facilitated domain adaptive semantic segmentation, which iteratively generates pseudo labels on the target domain and retrains the network. However, since the realistic segmentation datasets are highly imbalanced, target pseudo labels are typically biased to the majority classes and basically noisy, leading to an errorprone and sub-optimal model. To address this issue, we propose a region-based active learning approach for semantic segmentation under a domain shift, aiming to automatically query a small partition of image regions to be labeled while maximizing segmentation performance. Our algorithm, Active Learning via Region Impurity and Prediction Uncertainty (AL-RIPU), introduces a novel acquisition strategy characterizing the spatial adjacency of image regions along with the prediction confidence. We show that the proposed region-based selection strategy makes more efficient use of a limited budget than image-based or pointbased counterparts. Meanwhile, we enforce local prediction consistency between a pixel and its nearest neighbor on a source image. Further, we develop a negative learning loss to enhance the discriminative representation learning on the target domain. Extensive experiments demonstrate that our method only requires very few annotations to almost reach the supervised performance and substantially outperforms state-of-the-art methods.* Corresponding author (a) Target image (b) Image-based selection (100%) (c) Point-based selection (2.2%) (d) Region-based selection (2.2%)

show abstract

ReDAL: Region-based and Diversity-aware Active Learning for Point Cloud Semantic Segmentation

Cited by 55 publications

References 22 publications

Active Learning for Point Cloud Semantic Segmentation via Spatial-Structural Diversity Reasoning

Active Learning for Point Cloud Semantic Segmentation via Spatial-Structural Diversity Reasoning

LiDAR Dataset Distillation within Bayesian Active Learning Framework Understanding the Effect of Data Augmentation

Towards Fewer Annotations: Active Learning via Region Impurity and Prediction Uncertainty for Domain Adaptive Semantic Segmentation

Contact Info

Product

Resources

About