Non-parametric Memory for Spatio-Temporal Segmentation of Construction Zones for Self-Driving
Not on the map. Why?
Abstract (arXiv)
In this paper, we introduce a non-parametric memory representation for spatio-temporal segmentation that captures the local space and time around an autonomous vehicle (AV). Our representation has three important properties: (i) it remembers what it has seen in the past, (ii) it reinforces and (iii) forgets its past beliefs based on new evidence. Reinforcing is important as the first time we see an element we might be uncertain, e.g, if the element is heavily occluded or at range. Forgetting is desirable, as otherwise false positives will make the self driving vehicle behave erratically. Our process is informed by 3D reasoning, as occlusion is key to distinguishing between the desire to forget and to remember. We show how our method can be used as an online component to complement static world representations such as HD maps by detecting and remembering changes that should be superimposed on top of this static view due to such events.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 27 of its 45 references that are in the map. First posted on arXiv 18 Jan 2021.
Builds on
27 of 45 references.
References · 27
- 2015Deep Residual Learning for Image Recognitioncited by 51,809
- 2014Adam: A Method for Stochastic Optimizationcited by 36,960
- 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,223
- 2014Fully Convolutional Networks for Semantic Segmentationcited by 4,845
- 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,384
- 2016Pyramid Scene Parsing Networkcited by 2,472
- 2016Multi-View 3D Object Detection Network for Autonomous Drivingcited by 721
- 2017Frustum PointNets for 3D Object Detection from RGB-D Datacited by 640
- 2014Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFscited by 552
- 2014Predicting Depth, Surface Normals and Semantic Labels with a Common Multi-Scale Convolutional Architecturecited by 483
- 2016OctNet: Learning Deep 3D Representations at High Resolutionscited by 340
- 2017Joint 3D Proposal Generation and Object Detection from View Aggregationcited by 325
15 more
- 2020Deep Continuous Fusion for Multi-Sensor 3D Object Detectioncited by 263
- 2015Conditional Random Fields as Recurrent Neural Networkscited by 213
- 2018Tangent Convolutions for Dense Prediction in 3Dcited by 173
- 2021Deep Parametric Continuous Convolutional Neural Networkscited by 161
- 2018SPLATNet: Sparse Lattice Networks for Point Cloud Processingcited by 158
- 2016SemanticFusion: Dense 3D Semantic Mapping with Convolutional Neural Networkscited by 134
- 2016Vehicle Detection from 3D Lidar Using Fully Convolutional Networkcited by 105
- 2018Recurrent Slice Networks for 3D Segmentation of Point Cloudscited by 82
- 2017Learning Video Object Segmentation with Visual Memorycited by 66
- 2017Semantic Visual Localizationcited by 54
- 2018SBNet: Sparse Blocks Network for Fast Inferencecited by 53
- 2019Deep Multi-Sensor Lane Detectioncited by 30
- 2015Higher Order Conditional Random Fields in Deep Neural Networkscited by 25
- 2015Recurrent Instance Segmentationcited by 21
- 2015Parallel Multi-Dimensional LSTM, With Application to Fast Biomedical Volumetric Image Segmentationcited by 16
Built on by
Citing papers per year: 2023: 1.
1 paper.