Robust Multi-Modality Multi-Object Tracking

@article{Zhang2019RobustMM,
  title={Robust Multi-Modality Multi-Object Tracking},
  author={Wenwei Zhang and Hui Zhou and Shuyang Sun and Zhe Wang and Jianping Shi and Chen Change Loy},
  journal={2019 IEEE/CVF International Conference on Computer Vision (ICCV)},
  year={2019},
  pages={2365-2374}
}
Multi-sensor perception is crucial to ensure the reliability and accuracy in autonomous driving system, while multi-object tracking (MOT) improves that by tracing sequential movement of dynamic objects. Most current approaches for multi-sensor multi-object tracking are either lack of reliability by tightly relying on a single input source (e.g., center camera), or not accurate enough by fusing the results from multiple sensors in post processing without fully exploiting the inherent information… Expand
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
Modeling Cross-Modal Interaction in a Multi-detector, Multi-modal Tracking Framework
TLDR
The value and ability of the design in introducing a multi-modal tracking solution to the current research on Multi-Object Tracking (MOT) is proved. Expand
Joint Multi-Object Detection and Tracking with Camera-LiDAR Fusion for Autonomous Driving
  • Kemiao Huang, Qi Hao
  • Computer Science
  • ArXiv
  • 2021
TLDR
An efficient multi-modal MOT framework with online joint detection and tracking schemes and robust data association for autonomous driving applications and a comprehensive data association module for joint optimization among detection confidences, affinities and start-end probabilities is presented. Expand
Multi-modality Cascaded Fusion Technology for Autonomous Driving
TLDR
A general multi-modality cascaded fusion framework, exploiting the advantages of decision-level and feature-level fusion, utilizing target position, size, velocity, appearance and confidence to achieve accurate fusion results. Expand
Relation3DMOT: Exploiting Deep Affinity for 3D Multi-Object Tracking from View Aggregation
TLDR
This work proposes a novel convolutional operation, named RelationConv, to better exploit the correlation between each pair of objects in the adjacent frames and learn a deep affinity matrix for further data association in 3D multi-object tracking. Expand
SDVTracker: Real-Time Multi-Sensor Association and Tracking for Self-Driving Vehicles
TLDR
A model that jointly optimizes both association and state estimation with a novel loss, an algorithm for determining ground-truth supervision, and a training procedure is detail, showing this system significantly outperforms hand-engineered methods on a real-world urban driving dataset. Expand
CFTrack: Center-based Radar and Camera Fusion for 3D Multi-Object Tracking
TLDR
This work proposes an endto-end network for joint object detection and tracking based on radar and camera sensor fusion and utilizes a greedy algorithm for object association, making it very suitable for autonomous driving applications. Expand
Graph Neural Networks for 3D Multi-Object Tracking
TLDR
This work proposes a novel joint feature extractor to learn appearance and motion features from 2D and 3D space and achieves state-of-the-art 3D MOT performance. Expand
Segment as Points for Efficient Online Multi-Object Tracking and Segmentation
TLDR
This paper proposes a highly effective method for learning instance embeddings based on segments by converting the compact image representation to un-ordered 2D point cloud representation and builds a more challenging MOTS dataset named APOLLO MOTS with higher instance density. Expand
GNN3DMOT: Graph Neural Network for 3D Multi-Object Tracking with Multi-Feature Learning
TLDR
This work proposes two techniques to improve the discriminative feature learning for MOT by introducing a novel feature interaction mechanism by introducing the Graph Neural Network and proposes a novel joint feature extractor to learn appearance and motion features from 2D and 3D space simultaneously. Expand
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 62 REFERENCES
No Blind Spots: Full-Surround Multi-Object Tracking for Autonomous Vehicles Using Cameras and LiDARs
TLDR
A modular framework for tracking multiple objects (vehicles), capable of accepting object proposals from different sensor modalities and a variable number of sensors, to produce continuous object tracks, a generalization of the MDP framework for MOT proposed by Xiang et al. Expand
Mono-Camera 3D Multi-Object Tracking Using Deep Learning Detections and PMBM Filtering
TLDR
A multi-object tracking algorithm that takes an image as input and produces trajectories of detected objects in a world coordinate system by using a deep neural network trained to detect and estimate the distance to objects from a single input image. Expand
Online Multi-Object Tracking Using Joint Domain Information in Traffic Scenarios
TLDR
A novel tracking method that solves the problem of visual tracking of multiple objects by put together information from both enlarged structural and temporal domain by putting together the heterogeneous domain information, which exhibits an improved state-of-the-art performance on standard benchmarks. Expand
Occlusion Geodesics for Online Multi-object Tracking
TLDR
This work exploits the spatio-temporal evolution of occlusion regions, detector reliability, and target motion prediction to robustly handle missed detections and proposes an online approach based on the observation that object detectors primarily fail if objects are significantly occluded. Expand
Beyond Pixels: Leveraging Geometry and Shape Cues for Online Multi-Object Tracking
This paper introduces geometry and object shape and pose costs for multi-object tracking in urban driving scenarios. Using images from a monocular camera alone, we devise pairwise costs for objectExpand
Multi-Target Tracking using a 3D-Lidar sensor for autonomous vehicles
TLDR
This paper presents a robust algorithm for Multi-Target Tracking (MTT) using a Velodyne 3D HDL-64 Lidar sensor, and introduces hybrid ground classification and the Region of Interest (ROI) identification method in order to filter out the amount of unwanted raw data for the actual tracking. Expand
Multi-view 3D Object Detection Network for Autonomous Driving
TLDR
This paper proposes Multi-View 3D networks (MV3D), a sensory-fusion framework that takes both LIDAR point cloud and RGB images as input and predicts oriented 3D bounding boxes and designs a deep fusion scheme to combine region-wise features from multiple views and enable interactions between intermediate layers of different paths. Expand
Deep Continuous Conditional Random Fields With Asymmetric Inter-Object Constraints for Online Multi-Object Tracking
TLDR
A deep continuous conditional random field (DCCRF) is proposed for solving the online MOT problem in a track-by-detection framework and is trained in an end-to-end manner for better adapting the influences of visual information as well as inter-object relations. Expand
Learning to associate: HybridBoosted multi-target tracker for crowded scene
TLDR
A learning-based hierarchical approach of multi-target tracking from a single camera by progressively associating detection responses into longer and longer track fragments (tracklets) and finally the desired target trajectories by virtue of a HybridBoost algorithm. Expand
Online Multiperson Tracking-by-Detection from a Single, Uncalibrated Camera
TLDR
This paper proposes a novel approach for multiperson tracking-by-detection in a particle filtering framework that detects and tracks a large number of dynamically moving people in complex scenes with occlusions, requires no camera or ground plane calibration, and only makes use of information from the past. Expand
...
1
2
3
4
5
...