Deep Projective 3D Semantic Segmentation

By Felix Järemo Lawin, Martin Danelljan, Patrik Tosteberg, Goutam Bhat, Fahad Shahbaz Khan, Michael Felsberg. International Conference on Computer Analysis of Images and Patterns, 2017

361 citations recorded by the source corpus.

Semantic segmentation of 3D point clouds is a challenging problem with numerous real-world applications. While deep learning has revolutionized the field of image semantic segmentation, its impact on point cloud data has been limited so far. Recent attempts, based on 3D deep learning approaches (3D-CNNs), have achieved below-expected results. Such methods require voxelizations of the underlying point cloud data, leading to decreased spatial resolution and increased memory consumption. Additionally, 3D-CNNs greatly suffer from the limited availability of annotated datasets.

Subfields: Computational Mechanics, Environmental Engineering, Geology

View Paper: https://www.semanticscholar.org/paper/c538865c619113cb82b54e74646c32a8cafd6063

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Mechanics

In Environmental Engineering

In Geology

Citations in the corpus

Deep Learning for LiDAR Point Clouds in Autonomous Driving: A Review (2020, methodology): “Its main advantage is the use of stacked Conv layers with small receptive fields, which produces a lightweight network with limited parameters and increasing nonlinearity [25, 107, 115].”
SnapNet: 3D point cloud semantic labeling with 2D deep segmentation networks (2017, methodology): “DeePr3SS [39] is also a multi-view approach, very similar to ours but later, with multi-stream fully convolutional networks for 2D classiﬁcation of virtual views rendered by Gaussian point splattering.”
Revisiting Point Cloud Shape Classification with a Simple and Effective Baseline (2021, background): “(Lawin et al., 2017) render a scene point cloud from 120 views for different modalities like color, depth, and surface normal.”
PAConv: Position Adaptive Convolution with Dynamic Kernel Assembling on Point Clouds (2021, background): “Since point cloud data has irregular structure in 3D space, early works [44, 21, 6] project point clouds to multiview images and then utilize conventional convolutions for feature learning.”
KPConv: Flexible and Deformable Convolution for Point Clouds (2019, background): “Image-based networks are often multi-view, using a set of 2D images rendered from the point cloud at different viewpoints [34, 4, 17].”
Sparse Single Sweep LiDAR Point Cloud Segmentation via Learning Contextual Shape Priors from Scene Completion (2020, background): “Previous works projected all points scanned by the rotating LiDAR onto 2D images by plane projection (Lawin et al. 2017; Boulch, Le Saux, and Audebert 2017; Tatarchenko et al. 2018) or spherical projection (Wu et al. 2018, 2019).”
Deep Learning for 3D Point Clouds: A Survey (2019, methodology): “The first step of both the projection and discretizationbased methods is to transform a point cloud to an intermediate regular representation, such as multi-view [181], [182],”
ShellNet: Efficient Point Cloud Convolutional Neural Networks Using Concentric Shells Statistics (2019, neutral): “2 DeePr3SS [16] 58.”
SalsaNext: Fast, Uncertainty-Aware Semantic Segmentation of LiDAR Point Clouds (2020, methodology): “Each view is then processed by a multi-stream network as in [21].”
2DPASS: 2D Priors Assisted Semantic Segmentation on LiDAR Point Clouds (2022, background): “Previous works project all points scanned by the rotating LiDAR onto 2D images by plane projection [34–36], spherical projection [37,38] or both [39].”
SEGCloud: Semantic Segmentation of 3D Point Clouds (2017, neutral): “-made natural high low buildings hard scanning cars mIOU mAcc3 terrain terrain vegetation vegetation scape artefacts TMLC-MSR [27] 89.80 74.50 53.70 26.80 88.80 18.90 36.40 44.70 54.20 68.95 DeePr3SS [41] 85.60 83.20 74.20 32.40 89.70 18.50 25.10 59.20 58.50 88.90 SnapNet [6] 82.00 77.30 79.70 22.90”