MMM
YYYY
Instance Segmentation in 3D Scenes using Semantic Superpoint Tree Networks
基于语义重叠树网络的三维场景实例分割
意味的スーパーポイントツリーネットワークを用いた3 Dシーンにおけるインスタンスセグメンテーション
의미 중첩 트 리 네트워크 를 바탕 으로 하 는 3 차원 장면 인 스 턴 스 분할
Segmentación de instancias de escena 3D basada en la red de árboles superpuestos semánticos
Segmentation de l'Instance de scène 3D basée sur un réseau d'arbres de chevauchement sémantique
трехмерная модель, основанная на семантическом перекрытии дерева
Zhihao Liang ¹ ², Zhihao Li ³, Songcen Xu 许松岑 ³, Mingkui Tan 谭明奎 ¹, Kui Jia 贾奎 ¹ ⁴ ⁵
¹ South China University of Technology
华南理工大学
² DexForce Technology Co., Ltd.
跨维(广州)智能科技有限公司
³ Noah’s Ark Lab, Huawei Technologies
华为诺亚方舟实验室
⁴ Pazhou Laboratory
琶洲实验室 (人工智能与数字经济广东省实验室)
⁵ Peng Cheng Laboratory
鹏城实验室
arXiv, 17 August 2021
Abstract

Instance segmentation in 3D scenes is fundamental in many applications of scene understanding. It is yet challenging due to the compound factors of data irregularity and uncertainty in the numbers of instances. State-of-the-art methods largely rely on a general pipeline that first learns point-wise features discriminative at semantic and instance levels, followed by a separate step of point grouping for proposing object instances. While promising, they have the shortcomings that (1) the second step is not supervised by the main objective of instance segmentation, and (2) their point-wise feature learning and grouping are less effective to deal with data irregularities, possibly resulting in fragmented segmentations.

To address these issues, we propose in this work an end-to-end solution of Semantic Superpoint Tree Network (SSTNet) for proposing object instances from scene points. Key in SSTNet is an intermediate, semantic superpoint tree (SST), which is constructed based on the learned semantic features of superpoints, and which will be traversed and split at intermediate tree nodes for proposals of object instances. We also design in SSTNet a refinement module, termed CliqueNet, to prune superpoints that may be wrongly grouped into instance proposals.

Experiments on the benchmarks of ScanNet and S3DIS show the efficacy of our proposed method. At the time of submission, SSTNet ranks top on the ScanNet (V2) leaderboard, with 2% higher of mAP than the second best method.
arXiv_1
arXiv_2
arXiv_3
arXiv_4
Reviews and Discussions
https://www.hotpaper.io/index.html
Surrogate-guided conditional generative adversarial network for generative design of 3D-printable terahertz holographic metasurfaces
Statistical SERS spectroscopy based on concentration-encoded intensity distributions for deep-learning-assisted quantitation
Reconfigurable SERS platform based on magnetization-controlled self-assembly of magnetic nanoparticles on NdFeB composite substrate
Laser-induced biocompatible molybdenum dioxide for versatile wearable bioelectronics
Femtosecond laser inscribed 3D optical chirality
Topology transfer from vortex electrons to structured radiation via guided Cherenkov emission: orbital angular momentum and polarization skyrmions
Deep-learning-assisted label-free single-cell identification and sorting using femtosecond laser fabricated microfluidic surface-enhanced Raman scattering (SERS) chips with plasmonic ring-shaped nanostructure arrays
Ultrawide-angle diffraction-limited 2D beam steering via hybrid integrated metasurface-photonic circuit
Learning from hybrid bright-dark field imaging for resolution-enhanced digital pathology
Spectrally encoded diffractive optical unclonable functions using graphene zone plates
Coupled integrated photonic quantum memristors using a single photon source made of a colour center
Comprehensive review of distributed acoustic sensing technology: principles, configurations, applications and emerging trends



Previous Article                                Next Article
About
|
Contact
|
Copyright © Hot Paper