# DeepTemporalSeg: Temporally Consistent Semantic Segmentation of 3D LiDAR Scans

## TL;DR

This paper introduces DeepTemporalSeg, a deep neural network for semantic segmentation of 3D LiDAR scans that ensures temporal consistency using a Bayes filter, demonstrating improved accuracy on the KITTI benchmark.

## Contribution

The paper presents a novel DCNN architecture with dense blocks and depthwise convolutions for efficient semantic segmentation, combined with a Bayes filter for temporal consistency.

## Key findings

- Achieves state-of-the-art segmentation accuracy
- Improves temporal consistency of predictions
- Outperforms existing neural network architectures

## Abstract

Understanding the semantic characteristics of the environment is a key enabler for autonomous robot operation. In this paper, we propose a deep convolutional neural network (DCNN) for the semantic segmentation of a LiDAR scan into the classes car, pedestrian or bicyclist. This architecture is based on dense blocks and efficiently utilizes depth separable convolutions to limit the number of parameters while still maintaining state-of-the-art performance. To make the predictions from the DCNN temporally consistent, we propose a Bayes filter based method. This method uses the predictions from the neural network to recursively estimate the current semantic state of a point in a scan. This recursive estimation uses the knowledge gained from previous scans, thereby making the predictions temporally consistent and robust towards isolated erroneous predictions. We compare the performance of our proposed architecture with other state-of-the-art neural network architectures and report substantial improvement. For the proposed Bayes filter approach, we show results on various sequences in the KITTI tracking benchmark.