arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2404.04050cs.CV
收藏

无暇训练:赋能用于少样本3D场景分割的非参数化网络

No Time to Train: Empowering Non-Parametric Networks for Few-shot 3D Scene Segmentation

Shanghai AI Lab(上海人工智能实验室);City University of Hong Kong(香港城市大学);Peking University(北京大学);Tencent Youtu Lab(腾讯优图实验室);The Chinese University of Hong Kong(香港中文大学)

Xiangyang Zhu, Renrui Zhang, Bowei He, Ziyu Guo, Jiaming Liu, Han Xiao, Chaoyou Fu, Hao Dong, Peng Gao

AI总结:

本文提出免训练非参数化3D少样本分割网络Seg-NN及仅训练轻量QUEST模块的Seg-PN,在S3DIS与ScanNet上显著提升mIoU并减少90%训练时间。

摘要:

为减少对大规模数据集的依赖,近期3D分割工作采用少样本学习。当前3D少样本分割方法首先在“已见”类别上预训练模型,随后评估其在“未见”类别上的泛化性能。然而,先前的预训练阶段不仅引入过多时间开销,还会在“未见”类别上造成显著的域差距。为解决这些问题,我们提出一种用于少样本3D分割的非参数化网络Seg-NN,以及其参数化变体Seg-PN。Seg-NN无需训练,通过手工设计的滤波器提取稠密表示,并取得与现有参数化模型相当的性能。由于省去预训练,Seg-NN能够缓解域差距问题并节省大量时间。在Seg-NN基础上,Seg-PN仅需训练一个轻量级的QUEry-Support Transferring(QUERY-支持集迁移,QUEST)模块,该模块增强支持集与查询集之间的交互。实验表明,Seg-PN在S3DIS和ScanNet数据集上分别比此前最先进方法高出+4.19%和+7.71% mIoU,同时将训练时间减少-90%,表明其有效性和高效性。

英文摘要:

To reduce the reliance on large-scale datasets, recent works in 3D segmentation resort to few-shot learning. Current 3D few-shot segmentation methods first pre-train models on 'seen' classes, and then evaluate their generalization performance on 'unseen' classes. However, the prior pre-training stage not only introduces excessive time overhead but also incurs a significant domain gap on 'unseen' classes. To tackle these issues, we propose a Non-parametric Network for few-shot 3D Segmentation, Seg-NN, and its Parametric variant, Seg-PN. Without training, Seg-NN extracts dense representations by hand-crafted filters and achieves comparable performance to existing parametric models. Due to the elimination of pre-training, Seg-NN can alleviate the domain gap issue and save a substantial amount of time. Based on Seg-NN, Seg-PN only requires training a lightweight QUEry-Support Transferring (QUEST) module, which enhances the interaction between the support set and query set. Experiments suggest that Seg-PN outperforms previous state-of-the-art method by +4.19% and +7.71% mIoU on S3DIS and ScanNet datasets respectively, while reducing training time by -90%, indicating its effectiveness and efficiency.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑