SSPC-Net: Semi-supervised Semantic 3D Point Cloud Segmentation Network

Cheng, Mingmei; Hui, Le; Xie, Jin; Yang, Jian

doi:10.48550/arxiv.2104.07861

Cited by 5 publications

(5 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Semi-/Un Supervised Learning in Point Clouds. The semi-/un-supervised learning has been studied in various point cloud tasks, such as object detection [16], [17], semantic segmentation [43], [44] and scene flow estimation [45], [46]. Most of these methods first use the supervision from labeled data (either given or manually synthesized) to generate pseudo labels for unlabeled data.…”

Section: Related Workmentioning

confidence: 99%

Beyond 3D Siamese Tracking: A Motion-Centric Paradigm for 3D Single Object Tracking in Point Clouds

Zheng

Zhang

et al. 2022

2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

View full text Add to dashboard Cite

3D single object tracking in LiDAR point clouds (LiDAR SOT) plays a crucial role in autonomous driving. Current approaches all follow the Siamese paradigm based on appearance matching. However, LiDAR point clouds are usually textureless and incomplete, which hinders effective appearance matching. Besides, previous methods greatly overlook the critical motion clues among targets. In this work, beyond 3D Siamese tracking, we introduce a motion-centric paradigm to handle LiDAR SOT from a new perspective. Following this paradigm, we propose a matching-free two-stage tracker M 2 -Track. At the 1 st -stage, M 2 -Track localizes the target within successive frames via motion transformation. Then it refines the target box through motion-assisted shape completion at the 2 nd -stage. Due to the motion-centric nature, our method shows its impressive generalizability with limited training labels and provides good differentiability for end-to-end cycle training. This inspires us to explore semi-supervised LiDAR SOT by incorporating a pseudo-label-based motion augmentation and a self-supervised loss term. Under the fully-supervised setting, extensive experiments confirm that M 2 -Track significantly outperforms previous state-of-the-arts on three large-scale datasets while running at 57FPS (∼ 8%, ∼ 17% and ∼ 22% precision gains on KITTI, NuScenes, and Waymo Open Dataset respectively). While under the semi-supervised setting, our method performs on par with or even surpasses its fully-supervised counterpart using fewer than half labels from KITTI. Further analysis verifies each component's effectiveness and shows the motion-centric paradigm's promising potential for auto-labeling and unsupervised domain adaptation. The code is available at https://github.com/Ghostish/Open3DSOT.

show abstract

Section: Related Workmentioning

confidence: 99%

Beyond 3D Siamese Tracking: A Motion-Centric Paradigm for 3D Single Object Tracking in Point Clouds

Zheng

Zhang

et al. 2022

2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

View full text Add to dashboard Cite

show abstract

“…When tackling semisupervised segmentation on LiDAR point clouds, Semisup [18] implements a pseudo-label guided point contrastive loss to extend supervision to unlabeled frames. Li et al [23] and SSPC [10] employ self-training to achieve the same goal. Xu et al [52] compares semi-supervised training to weakly-supervised on point clouds and argues that under a fixed labelling budget, weak supervision performs better for semantic segmentation.…”

Section: Incomplete Supervision In 3d Semantic Segmentationmentioning

confidence: 99%

“…WS3D [20] utilizes region-level boundary awareness and instance discrimination to improve indoor and outdoor 3D semantic segmentation with simulated weak labels. Furthermore for semi-supervised learning, DiAL [32] uses a simple MT setup, GPC [16] proposes using a pseudo-label guided point contrastive loss, SSPC [8] utilizes self-training and LaserMix [17] uses a mixing operation to bring supervision to unlabeled frames. CPS [7] utilizes a Siamese structure to induce cross supervision.…”

Section: Related Workmentioning

confidence: 99%

Scribble-Supervised LiDAR Semantic Segmentation

Unal

Dai

Gool

2022

2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

View full text Add to dashboard Cite

Densely annotating LiDAR point clouds remains too expensive and time-consuming to keep up with the ever growing volume of data. While current literature focuses on fully-supervised performance, developing efficient methods that take advantage of realistic weak supervision have yet to be explored. In this paper, we propose using scribbles to annotate LiDAR point clouds and release ScribbleKITTI, the first scribble-annotated dataset for LiDAR semantic segmentation. Furthermore, we present a pipeline to reduce the performance gap that arises when using such weak annotations. Our pipeline comprises of three stand-alone contributions that can be combined with any LiDAR semantic segmentation model to achieve up to 95.7% of the fully-supervised performance while using only 8% labeled points. Our scribble annotations and code are available at github.com/ouenal/scribblekitti.

show abstract

“…WS3D [19] utilizes region-level boundary awareness and instance discrimination to improve indoor and outdoor 3D semantic segmentation with simulated weak labels. Furthermore for semi-supervised learning, DiAL [31] uses a simple MT setup, GPC [15] proposes using a pseudo-label guided point contrastive loss, SSPC [8] utilizes self-training and LaserMix [16] uses a mixing operation to bring supervision to unlabeled frames. CPS [7] utilizes a Siamese structure to induce cross supervision.…”

Section: Related Workmentioning

confidence: 99%

2D Feature Distillation for Weakly- and Semi-Supervised 3D Semantic Segmentation

Unal,

Dai,

Hoyer

et al. 2024

2024 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)

View full text Add to dashboard Cite

As 3D perception problems grow in popularity and the need for large-scale labeled datasets for LiDAR semantic segmentation increase, new methods arise that aim to reduce the necessity for dense annotations by employing weakly-supervised training. However these methods continue to show weak boundary estimation and high false negative rates for small objects and distant sparse regions. We argue that such weaknesses can be compensated by using RGB images which provide a denser representation of the scene. We propose an image-guidance network (IGNet) which builds upon the idea of distilling high level feature information from a domain adapted synthetically trained 2D semantic segmentation network. We further utilize a oneway contrastive learning scheme alongside a novel mixing strategy called FOVMix, to combat the horizontal field-ofview mismatch between the two sensors and enhance the effects of image guidance. IGNet achieves state-of-the-art results for weakly-supervised LiDAR semantic segmentation on ScribbleKITTI, boasting up to 98% relative performance to fully supervised training with only 8% labeled points, while introducing no additional annotation burden or computational/memory cost during inference. Furthermore, we show that our contributions also prove effective for semisupervised training, where IGNet claims state-of-the-art results on both ScribbleKITTI and SemanticKITTI.

show abstract

SSPC-Net: Semi-supervised Semantic 3D Point Cloud Segmentation Network

Cited by 5 publications

References 0 publications

Beyond 3D Siamese Tracking: A Motion-Centric Paradigm for 3D Single Object Tracking in Point Clouds

Beyond 3D Siamese Tracking: A Motion-Centric Paradigm for 3D Single Object Tracking in Point Clouds

Scribble-Supervised LiDAR Semantic Segmentation

2D Feature Distillation for Weakly- and Semi-Supervised 3D Semantic Segmentation

Contact Info

Product

Resources

About