3D Multi-Object Tracking: A Baseline and New Evaluation Metrics

@article{Weng20203DMT,
  title={3D Multi-Object Tracking: A Baseline and New Evaluation Metrics},
  author={Xinshuo Weng and Jianren Wang and David Held and Kris Kitani},
  journal={2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
  year={2020},
  pages={10359-10366}
}
3D multi-object tracking (MOT) is an essential component for many applications such as autonomous driving and assistive robotics. Recent work on 3D MOT focuses on developing accurate systems giving less attention to practical considerations such as computational cost and system complexity. In contrast, this work proposes a simple real-time 3D MOT system. Our system first obtains 3D detections from a LiDAR point cloud. Then, a straightforward combination of a 3D Kalman filter and the Hungarian… Expand

Figures and Tables from this paper

EagerMOT: 3D Multi-Object Tracking via Sensor Fusion
TLDR
This paper proposes EagerMOT, a simple tracking formulation that eagerly integrates all available object observations from both sensor modalities to obtain a well-informed interpretation of the scene dynamics and achieves state-of-the-art results across several MOT tasks on the KITTI and NuScenes datasets. Expand
4D Panoptic LiDAR Segmentation
TLDR
This paper proposes 4D panoptic LiDAR segmentation to assign a semantic class and a temporallyconsistent instance ID to a sequence of 3D points and proposes a new evaluation metric that separates the semantic and point-to-instance association aspects of the task. Expand
Tracklet Proposal Network for Multi-Object Tracking on Point Clouds
  • Hai Wu, Qing Li, Chenglu Wen, Xin Li, Xiaoliang Fan, Cheng Wang
  • Computer Science
  • IJCAI
  • 2021
This paper proposes the first tracklet proposal network, named PC-TCNN, for Multi-Object Tracking (MOT) on point clouds. Our pipeline first generates tracklet proposals, then refines these trackletsExpand
FlowMOT: 3D Multi-Object Tracking by Scene Flow Association
TLDR
A LiDAR-based 3D MOT framework named FlowMOT is proposed, which integrates point-wise motion information into the traditional matching algorithm, enhancing the robustness of the data association and outperforms recent end-to-end methods. Expand
Lightweight Map-Enhanced 3D Object Detection and Tracking for Autonomous Driving
  • Lei Gong, Shunhong Wang, Yu Zhang, Yanyong Zhang, Jianmin Ji
  • Computer Science
  • Internetware
  • 2020
TLDR
A holistic approach that integrates and utilizes the map information in system following the tracking-by-detection paradigm is proposed, and it is shown that this framework can improve the detection and tracking accuracy with mild or no increase of latency. Expand
Visual-LiDAR Based 3D Object Detection and Tracking for Embedded Systems
TLDR
The experiments suggest that the proposed framework offers a suitable solution for embedded systems to solve 3D object detection and tracking problem, with added benefits. Expand
DiTNet: End-to-End 3D Object Detection and Track ID Assignment in Spatio-Temporal World
TLDR
An end-to-end point cloud-based network, DiTNet, to directly assign a track ID to each object across the whole sequence, without the data association step, achieves competitive performance over the state-of-the-art methods on the KITTI benchmark. Expand
Driving Behavior-Aware Network for 3D Object Tracking in Complex Traffic Scenes
TLDR
This work presents an end-to-end deep learning framework with a driving behavior-aware model that takes full advantage of spatial-temporal details in consecutive frames and learns the driving behavior from object variations in 2D center point, depth, rotation and translation in parallel. Expand
Learnable Online Graph Representations for 3D Multi-Object Tracking
TLDR
This work designs a graph structure to jointly process detection and track states in an online manner and employs a Neural Message Passing network for data association that is fully trainable, providing a natural way for track initialization and handling of false positive detections, while significantly improving track stability. Expand
Navigate-and-Seek: A Robotics Framework for People Localization in Agricultural Environments
TLDR
By combining multi-sensor observations on the topological level complemented by active perception through the NBS approach, it is shown that the accuracy of picker localization in comparison to prior work can be improved. Expand
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 59 REFERENCES
Learning Optimal Parameters for Multi-target Tracking with Contextual Interactions
TLDR
Surprisingly, with proper parameter learning, the simple data association model without explicit appearance/motion reasoning is able to achieve comparable or better accuracy than many state-of-the-art methods that use far more complex motion features or appearance affinity metric learning. Expand
Multi-Class Multi-Object Tracking using Changing Point Detection
TLDR
This paper presents a robust multi-class multi-object tracking (MCMOT) formulated by a Bayesian filtering framework and comparison to state-of-the-art video tracking techniques shows very encouraging results. Expand
FANTrack: 3D Multi-Object Tracking with Feature Association Network
TLDR
A data-driven approach to online multi-object tracking that uses a convolutional neural network for data association in a tracking-by-detection framework that learns to perform global assignments in 3D purely from data, handles noisy detections and varying number of targets, and is easy to train. Expand
Focal Loss for Dense Object Detection
TLDR
This paper proposes to address the extreme foreground-background class imbalance encountered during training of dense detectors by reshaping the standard cross entropy loss such that it down-weights the loss assigned to well-classified examples, and develops a novel Focal Loss, which focuses training on a sparse set of hard examples and prevents the vast number of easy negatives from overwhelming the detector during training. Expand
PointRCNN: 3D Object Proposal Generation and Detection From Point Cloud
TLDR
Extensive experiments on the 3D detection benchmark of KITTI dataset show that the proposed architecture outperforms state-of-the-art methods with remarkable margins by using only point cloud as input. Expand
Robust Multi-Modality Multi-Object Tracking
TLDR
This study designs a generic sensor-agnostic multi-modality MOT framework (mmMOT), where each modality is capable of performing its role independently to preserve reliability, and could further improving its accuracy through a novel multi- modality fusion module. Expand
A Lightweight Online Multiple Object Vehicle Tracking Method
TLDR
This study is particularly focused on the development and benchmarking of an computationally lightweight online multiple object tracking model and its model outperforms the state-of-the-art methods on ML and MT and places second on MOTA and MOTP metric evaluations. Expand
HDNET: Exploiting HD Maps for 3D Object Detection
TLDR
It is shown that High-Definition maps provide strong priors that can boost the performance and robustness of modern 3D object detectors and a single stage detector is designed that extracts geometric and semantic features from the HD maps. Expand
Simple online and realtime tracking with a deep association metric
TLDR
This paper integrates appearance information to improve the performance of SORT and reduces the number of identity switches, achieving overall competitive performance at high frame rates. Expand
Simple online and realtime tracking
TLDR
Despite only using a rudimentary combination of familiar techniques such as the Kalman Filter and Hungarian algorithm for the tracking components, this approach achieves an accuracy comparable to state-of-the-art online trackers. Expand
...
1
2
3
4
5
...