An Enhanced SqueezeNet Based Network for Real-Time Road-Object Segmentation

Point cloud image segmentation plays an important role in self-driving. SqueezeSeg network has good performance in terms of accuracy and calculation speed on point cloud segmentation. However, potential details might be lost during the computational processing of SqueezeSeg and other similar kinds of networks. In this work, we try to retain the detailed information of the image by combining PointSeg network and the conditional random field in order to capture more data information and improve the recall rate. These two processes can complement and fully play their respective advantages. The proposed method has been tested on KITTI dataset. Simulation results demonstrate that our method can overcome the shortcomings of the SqueezeSeg network and similar kinds of networks on the extraction of detailed information.

[1]  Amnon Shashua,et al.  On a Formal Model of Safe and Scalable Self-driving Cars , 2017, ArXiv.

[2]  Vladlen Koltun,et al.  Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials , 2011, NIPS.

[3]  Forrest N. Iandola,et al.  SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <1MB model size , 2016, ArXiv.

[4]  Yuan Wang,et al.  PointSeg: Real-Time Semantic Segmentation Based on 3D LiDAR Point Cloud , 2018, ArXiv.

[5]  Tsuyoshi Murata,et al.  {m , 1934, ACML.

[6]  Vladlen Koltun,et al.  Multi-Scale Context Aggregation by Dilated Convolutions , 2015, ICLR.

[7]  Danna Zhou,et al.  d. , 1934, Microbial pathogenesis.

[8]  Andreas Geiger,et al.  Vision meets robotics: The KITTI dataset , 2013, Int. J. Robotics Res..

[9]  Yin Zhou,et al.  VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detection , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[10]  Vibhav Vineet,et al.  Conditional Random Fields as Recurrent Neural Networks , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[11]  Xuan Tang,et al.  Reconstructible Nonlinear Dimensionality Reduction via Joint Dictionary Learning , 2019, IEEE Transactions on Neural Networks and Learning Systems.

[12]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[13]  Wei Wu,et al.  PointCNN: Convolution On X-Transformed Points , 2018, NeurIPS.

[14]  Iasonas Kokkinos,et al.  DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs , 2016, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[15]  Huiming Qu,et al.  Research on Semantic Segmentation of High-resolution Remote Sensing Image Based on Full Convolutional Neural Network , 2018, 2018 12th International Symposium on Antennas, Propagation and EM Theory (ISAPE).

[16]  Aaas News,et al.  Book Reviews , 1893, Buffalo Medical and Surgical Journal.

[17]  Trevor Darrell,et al.  Fully Convolutional Networks for Semantic Segmentation , 2017, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[18]  Shinpei Kato,et al.  LMNet: Real-time Multiclass Object Detection on CPU Using 3D LiDAR , 2018, 2018 3rd Asia-Pacific Conference on Intelligent Robot Systems (ACIRS).

[19]  G. G. Stokes.  "J." , 1890, The New Yale Book of Quotations.

[20]  Enhua Wu,et al.  Squeeze-and-Excitation Networks , 2017, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[21]  Kurt Keutzer,et al.  SqueezeSeg: Convolutional Neural Nets with Recurrent CRF for Real-Time Road-Object Segmentation from 3D LiDAR Point Cloud , 2017, 2018 IEEE International Conference on Robotics and Automation (ICRA).

[22]  Iasonas Kokkinos,et al.  Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs , 2014, ICLR.

[23]  Leonidas J. Guibas,et al.  PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).