论文标题
salsanext：快速，不确定性感知的语义分段，用于自动驾驶的激光点云
SalsaNext: Fast, Uncertainty-aware Semantic Segmentation of LiDAR Point Clouds for Autonomous Driving
论文作者
论文摘要
在本文中，我们介绍了Salsanext，以实时对完整的3D激光点云进行不确定性感知的语义分割。 Salsanext是Salsanet [1]的下一个版本，该版本具有一个编码器架构，其中编码器单元具有一组重新网络块，而解码器部件结合了从残差块中的UPS采样功能。与Salsanet相反，我们引入了一个新的上下文模块，用新的残留扩张卷积堆栈替换Resnet编码器块，并逐渐增加接受场，并在解码器中添加Pixel-Shuffle层。此外，我们从Stride卷积切换到平均合并，还采用了中央辍学处理。为了直接优化Jaccard索引，我们将加权交叉渗透损失与Lovasz-Softmax损失相结合[2]。最终，我们注入了贝叶斯治疗，以计算云中每个点的认知和核心不确定性。我们对语义-KITTI数据集进行了彻底的定量评估[3]，该评估表明，所提出的Salsanext优于其他最先进的语义细分网络，并在语义Kitti排行榜上排名第一。我们还发布源代码https://github.com/tiagocortinhal/salsanext。
In this paper, we introduce SalsaNext for the uncertainty-aware semantic segmentation of a full 3D LiDAR point cloud in real-time. SalsaNext is the next version of SalsaNet [1] which has an encoder-decoder architecture where the encoder unit has a set of ResNet blocks and the decoder part combines upsampled features from the residual blocks. In contrast to SalsaNet, we introduce a new context module, replace the ResNet encoder blocks with a new residual dilated convolution stack with gradually increasing receptive fields and add the pixel-shuffle layer in the decoder. Additionally, we switch from stride convolution to average pooling and also apply central dropout treatment. To directly optimize the Jaccard index, we further combine the weighted cross-entropy loss with Lovasz-Softmax loss [2]. We finally inject a Bayesian treatment to compute the epistemic and aleatoric uncertainties for each point in the cloud. We provide a thorough quantitative evaluation on the Semantic-KITTI dataset [3], which demonstrates that the proposed SalsaNext outperforms other state-of-the-art semantic segmentation networks and ranks first on the Semantic-KITTI leaderboard. We also release our source code https://github.com/TiagoCortinhal/SalsaNext.
