# Learning 3D Semantic Segmentation with only 2D Image Supervision

arXiv: https://arxiv.org/abs/2110.11325 · HTML: https://mlsanity.com/paper/2110.11325/

## Key facts

- Cited by 60 papers in the map (data through 14 Oct 2026).
- Builds on 60 of its 99 references that are in the map.
- First posted on arXiv 21 Oct 2021.

## Builds on (most cited first, 30 of 60)

- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation](https://mlsanity.com/paper/1612.00593/) (2016)
- [nuScenes: A multimodal dataset for autonomous driving](https://mlsanity.com/paper/1903.11027/) (2019)
- [The Cityscapes Dataset for Semantic Urban Scene Understanding](https://mlsanity.com/paper/1604.01685/) (2016)
- [PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space](https://mlsanity.com/paper/1706.02413/) (2017)
- [ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes](https://mlsanity.com/paper/1702.04405/) (2017)
- [Xception: Deep Learning with Depthwise Separable Convolutions](https://mlsanity.com/paper/1610.02357/) (2016)
- [Rethinking Atrous Convolution for Semantic Image Segmentation](https://mlsanity.com/paper/1706.05587/) (2017)
- [Contrastive Multiview Coding](https://mlsanity.com/paper/1906.05849/) (2019)
- [4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks](https://mlsanity.com/paper/1904.08755/) (2019)
- [KPConv: Flexible and Deformable Convolution for Point Clouds](https://mlsanity.com/paper/1904.08889/) (2019)
- [PV-RCNN: Point-Voxel Feature Set Abstraction for 3D Object Detection](https://mlsanity.com/paper/1912.13192/) (2019)
- [3D Semantic Segmentation with Submanifold Sparse Convolutional Networks](https://mlsanity.com/paper/1711.10275/) (2017)
- [Deep Learning for 3D Point Clouds: A Survey](https://mlsanity.com/paper/1912.12033/) (2019)
- [Semantic Scene Completion from a Single Depth Image](https://mlsanity.com/paper/1611.08974/) (2016)
- [PointPainting: Sequential Fusion for 3D Object Detection](https://mlsanity.com/paper/1911.10150/) (2019)
- [Searching Efficient 3D Architectures with Sparse Point-Voxel Convolution](https://mlsanity.com/paper/2007.16100/) (2020)
- [Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR Segmentation](https://mlsanity.com/paper/2011.10033/) (2020)
- [OctNet: Learning Deep 3D Representations at High Resolutions](https://mlsanity.com/paper/1611.05009/) (2016)
- [3D Scene Graph: A Structure for Unified Semantics, 3D Space, and Camera](https://mlsanity.com/paper/1910.02527/) (2019)
- [PolarNet: An Improved Grid Representation for Online LiDAR Point Clouds Semantic Segmentation](https://mlsanity.com/paper/2003.14032/) (2020)
- [Self-Supervised Learning by Cross-Modal Audio-Video Clustering](https://mlsanity.com/paper/1911.12667/) (2019)
- [The Sound of Pixels](https://mlsanity.com/paper/1804.03160/) (2018)
- [Cooperative Learning of Audio and Video Models from Self-Supervised Synchronization](https://mlsanity.com/paper/1807.00230/) (2018)
- [SoundNet: Learning Sound Representations from Unlabeled Video](https://mlsanity.com/paper/1610.09001/) (2016)
- [SalsaNext: Fast, Uncertainty-aware Semantic Segmentation of LiDAR Point Clouds for Autonomous Driving](https://mlsanity.com/paper/2003.03653/) (2020)
- [OccuSeg: Occupancy-aware 3D Instance Segmentation](https://mlsanity.com/paper/2003.06537/) (2020)
- [Cross Modal Distillation for Supervision Transfer](https://mlsanity.com/paper/1507.00448/) (2015)
- [SqueezeSegV3: Spatially-Adaptive Convolution for Efficient Point-Cloud Segmentation](https://mlsanity.com/paper/2004.01803/) (2020)

## Built on by (most cited first, 30 of 60)

- [OpenScene: 3D Scene Understanding with Open Vocabularies](https://mlsanity.com/paper/2211.15654/) (2022)
- [2DPASS: 2D Priors Assisted Semantic Segmentation on LiDAR Point Clouds](https://mlsanity.com/paper/2207.04397/) (2022)
- [Rethinking Range View Representation for LiDAR Segmentation](https://mlsanity.com/paper/2303.05367/) (2023)
- [Image-to-Lidar Self-Supervised Distillation for Autonomous Driving Data](https://mlsanity.com/paper/2203.16258/) (2022)
- [Spherical Transformer for LiDAR-based 3D Recognition](https://mlsanity.com/paper/2303.12766/) (2023)
- [Delving into the Devils of Bird's-eye-view Perception: A Review, Evaluation and Recipe](https://mlsanity.com/paper/2209.05324/) (2022)
- [NeSF: Neural Semantic Fields for Generalizable Semantic Segmentation of 3D Scenes](https://mlsanity.com/paper/2111.13260/) (2021)
- [MSeg3D: Multi-modal 3D Semantic Segmentation for Autonomous Driving](https://mlsanity.com/paper/2303.08600/) (2023)
- [UniSeg: A Unified Multi-Modal LiDAR Segmentation Network and the OpenPCSeg Codebase](https://mlsanity.com/paper/2309.05573/) (2023)
- [FRNet: Frustum-Range Networks for Scalable LiDAR Segmentation](https://mlsanity.com/paper/2312.04484/) (2023)
- [Nerflets: Local Radiance Fields for Efficient Structure-Aware 3D Scene Representation from 2D Supervision](https://mlsanity.com/paper/2303.03361/) (2023)
- [SAMPro3D: Locating SAM Prompts in 3D for Zero-Shot Instance Segmentation](https://mlsanity.com/paper/2311.17707/) (2023)
- [SNAP: Self-Supervised Neural Maps for Visual Positioning and Semantic Understanding](https://mlsanity.com/paper/2306.05407/) (2023)
- [PartDistill: 3D Shape Part Segmentation by Vision-Language Model Distillation](https://mlsanity.com/paper/2312.04016/) (2023)
- [ODIN: A Single Model for 2D and 3D Segmentation](https://mlsanity.com/paper/2401.02416/) (2024)
- [GECCO: Geometrically-Conditioned Point Diffusion Models](https://mlsanity.com/paper/2303.05916/) (2023)
- [Lifting by Gaussians: A Simple, Fast and Flexible Method for 3D Instance Segmentation](https://mlsanity.com/paper/2502.00173/) (2025)
- [Contrastive Gaussian Clustering: Weakly Supervised 3D Scene Segmentation](https://mlsanity.com/paper/2404.12784/) (2024)
- [2D-3D Interlaced Transformer for Point Cloud Segmentation with Scene-Level Supervision](https://mlsanity.com/paper/2310.12817/) (2023)
- [LSK3DNet: Towards Effective and Efficient 3D Perception with Large Sparse Kernels](https://mlsanity.com/paper/2403.15173/) (2024)
- [Spot-Compose: A Framework for Open-Vocabulary Object Retrieval and Drawer Manipulation in Point Clouds](https://mlsanity.com/paper/2404.12440/) (2024)
- [RAPiD-Seg: Range-Aware Pointwise Distance Distribution Networks for 3D LiDAR Segmentation](https://mlsanity.com/paper/2407.10159/) (2024)
- [Neural Semantic Surface Maps](https://mlsanity.com/paper/2309.04836/) (2023)
- [TASeg: Temporal Aggregation Network for LiDAR Semantic Segmentation](https://mlsanity.com/paper/2407.09751/) (2024)
- [SFPNet: Sparse Focal Point Network for Semantic Segmentation on General LiDAR Point Clouds](https://mlsanity.com/paper/2407.11569/) (2024)
- [360^circ from a Single Camera: A Few-Shot Approach for LiDAR Segmentation](https://mlsanity.com/paper/2309.06197/) (2023)
- [3D Weakly Supervised Semantic Segmentation with 2D Vision-Language Guidance](https://mlsanity.com/paper/2407.09826/) (2024)
- [Data Efficient 3D Learner via Knowledge Transferred from 2D Model](https://mlsanity.com/paper/2203.08479/) (2022)
- [MvDeCor: Multi-view Dense Correspondence Learning for Fine-grained 3D Segmentation](https://mlsanity.com/paper/2208.08580/) (2022)
- [Mx2M: Masked Cross-Modality Modeling in Domain Adaptation for 3D Semantic Segmentation](https://mlsanity.com/paper/2307.04231/) (2023)

## Abstract (arXiv)

With the recent growth of urban mapping and autonomous driving efforts, there has been an explosion of raw 3D data collected from terrestrial platforms with lidar scanners and color cameras. However, due to high labeling costs, ground-truth 3D semantic segmentation annotations are limited in both quantity and geographic diversity, while also being difficult to transfer across sensors. In contrast, large image collections with ground-truth semantic segmentations are readily available for diverse sets of scenes. In this paper, we investigate how to use only those labeled 2D image collections to supervise training 3D semantic segmentation models. Our approach is to train a 3D model from pseudo-labels derived from 2D semantic image segmentations using multiview fusion. We address several novel issues with this approach, including how to select trusted pseudo-labels, how to sample 3D scenes with rare object categories, and how to decouple input features from 2D images from pseudo-labels during training. The proposed network architecture, 2D3DNet, achieves significantly better performance (+6.2-11.4 mIoU) than baselines during experiments on a new urban dataset with lidar and images captured in 20 cities across 5 continents.