Input-Output Balanced Framework for Long-Tailed Lidar Semantic Segmentation

@article{Cong2021InputOutputBF,
  title={Input-Output Balanced Framework for Long-Tailed Lidar Semantic Segmentation},
  author={Peishan Cong and Xinge Zhu and Yuexin Ma},
  journal={2021 IEEE International Conference on Multimedia and Expo (ICME)},
  year={2021},
  pages={1-6}
}
A thorough and holistic scene understanding is crucial for autonomous vehicles, where LiDAR semantic segmentation plays an indispensable role. However, most existing methods focus on the network design while neglecting the inherent difficulty, i.e, imbalanced data distribution in the realistic dataset (also named long-tailed distribution), which narrows down the capability of state-of-the-art methods. In this paper, we propose an input-output balanced framework to handle the issue of long… 

Figures and Tables from this paper

Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR-based Perception
  • Xinge Zhu, Hui Zhou, Da Lin
  • Computer Science
    IEEE transactions on pattern analysis and machine intelligence
  • 2021
TLDR
A new framework for the outdoor LiDAR segmentation is proposed, where cylindrical partition and asymmetrical 3D convolution networks are designed to explore the 3D geometric pattern while maintaining these inherent properties.
Self-supervised Point Cloud Completion on Real Traffic Scenes via Scene-concerned Bottom-up Mechanism
TLDR
This paper proposes a selfsupervised point cloud completion method (TraPCC) for vehicles in real traffic scenes without any complete data, based on the symmetry and similarity of vehicles, and designs a bottom-up mechanism to focus on both local geometry details and global shape features of inputs.
FusionPainting: Multimodal Fusion with Adaptive Attention for 3D Object Detection
TLDR
A general multimodal fusion framework to fuse the 2D RGB image and 3D point clouds at a semantic level for boosting the 3D object detection task and outperforms other state-of-the-art methods on the nuScenes testing benchmark.

References

SHOWING 1-10 OF 30 REFERENCES
SqueezeSegV2: Improved Model Structure and Unsupervised Domain Adaptation for Road-Object Segmentation from a LiDAR Point Cloud
TLDR
This work introduces a new model SqueezeSegV2, which is more robust against dropout noises in LiDAR point cloud and therefore achieves significant accuracy improvement, and a domain-adaptation training pipeline consisting of three major components: learned intensity rendering, geodesic correlation alignment, and progressive domain calibration.
Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR Segmentation
  • Xinge Zhu, Hui Zhou, Dahua Lin
  • Computer Science
    2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
  • 2021
TLDR
A new framework for the outdoor LiDAR segmentation is proposed, where cylindrical partition and asymmetrical 3D convolution networks are designed to explore the 3D geometric pattern while maintaining these inherent properties.
SalsaNext: Fast Semantic Segmentation of LiDAR Point Clouds for Autonomous Driving
TLDR
This paper provides a thorough quantitative evaluation on the Semantic-KITTI dataset, which demonstrates that the proposed SalsaNext outperforms other state-of-the-art semantic segmentation networks in terms of accuracy and computation time.
PolarNet: An Improved Grid Representation for Online LiDAR Point Clouds Semantic Segmentation
TLDR
This work proposes a new LiDAR-specific, KNN-free segmentation algorithm - PolarNet, which greatly increases the mIoU in three drastically different real urban LiDar single-scan segmentation datasets while retaining ultra low latency and near real-time throughput.
RangeNet ++: Fast and Accurate LiDAR Semantic Segmentation
TLDR
This paper proposes a novel post-processing algorithm that deals with problems arising from this intermediate representation of range images as an intermediate representation in combination with a Convolutional Neural Network exploiting the rotating LiDAR sensor model.
SemanticKITTI: A Dataset for Semantic Scene Understanding of LiDAR Sequences
TLDR
A large dataset to propel research on laser-based semantic segmentation, which opens the door for the development of more advanced methods, but also provides plentiful data to investigate new research directions.
Cylinder3D: An Effective 3D Framework for Driving-scene LiDAR Semantic Segmentation
TLDR
This work develops a 3D cylinder partition and a3D cylinder convolution based framework, termed as Cylinder3D, which exploits the 3D topology relations and structures of driving-scene point clouds and introduces a dimension-decomposition based context modeling module to explore the high-rank context information in point clouds in a progressive manner.
LiDAR-based Panoptic Segmentation via Dynamic Shifting Network
TLDR
This work proposes the Dynamic Shifting Network (DS-Net), which serves as an effective panoptic segmentation framework in the point cloud realm and presents an efficient learnable clustering module, dynamic shifting, which adapts kernel functions on-the-fly for different instances.
SEGCloud: Semantic Segmentation of 3D Point Clouds
TLDR
SEGCloud is presented, an end-to-end framework to obtain 3D point-level segmentation that combines the advantages of NNs, trilinear interpolation(TI) and fully connected Conditional Random Fields (FC-CRF).
SqueezeSeg: Convolutional Neural Nets with Recurrent CRF for Real-Time Road-Object Segmentation from 3D LiDAR Point Cloud
TLDR
An end-to-end pipeline called SqueezeSeg based on convolutional neural networks (CNN), which takes a transformed LiDAR point cloud as input and directly outputs a point-wise label map, which is then refined by a conditional random field (CRF) implemented as a recurrent layer.
...
1
2
3
...