Deep Learning in Video Multi-Object Tracking: A Survey

@article{Ciaparrone2020DeepLI,
  title={Deep Learning in Video Multi-Object Tracking: A Survey},
  author={Gioele Ciaparrone and Francisco Luque S{\'a}nchez and Siham Tabik and Luigi Troiano and Roberto Tagliaferri and Francisco Herrera},
  journal={Neurocomputing},
  year={2020},
  volume={381},
  pages={61-88}
}
The problem of Multiple Object Tracking (MOT) consists in following the trajectory of different objects in a sequence, usually a video. [...] Key Method Four main steps in MOT algorithms are identified, and an in-depth review of how Deep Learning was employed in each one of these stages is presented. A complete experimental comparison of the presented works on the three MOTChallenge datasets is also provided, identifying a number of similarities among the top-performing methods and presenting some possible…Expand
Multiple Object Tracking in Deep Learning Approaches: A Survey
  • Yesul Park, L. Dang, Sujin Lee, Dongil Han, Hyeonjoon Moon
  • Computer Science
  • Electronics
  • 2021
Object tracking is a fundamental computer vision problem that refers to a set of methods proposed to precisely track the motion trajectory of an object in a video. Multiple Object Tracking (MOT) is aExpand
Multi-target tracking for video surveillance using deep affinity network: a brief review
TLDR
The state-of-the-art MTT models, which leverage from deep learning representational power are reviewed, which show promising results in multi-target tracking for video surveillance. Expand
Real-time multiple object tracking using deep learning methods
TLDR
A real-time multiple-object tracking framework that is based on a modified version of the Deep SORT algorithm based on the process of the initialization of the objects, and its rationale is to consider an object as tracked if it is detected in a set of previous frames. Expand
Collaborative Multi-Object Tracking as an Edge Service using Transfer Learning
  • Han Sun, Yu Chen, Alexander J. Aved, Erik Blasch
  • Computer Science
  • 2020 IEEE 22nd International Conference on High Performance Computing and Communications; IEEE 18th International Conference on Smart City; IEEE 6th International Conference on Data Science and Systems (HPCC/SmartCity/DSS)
  • 2020
TLDR
Results verify that the CoMOTE prototype achieved the design goal and outperforms multiple state-of-the-art methods and leads to a lower communication cost. Expand
Learning a Proposal Classifier for Multiple Object Tracking
TLDR
A novel proposal-based learnable framework, which models MOT as a proposal generation, proposal scoring and trajectory inference paradigm on an affinity graph, and can solve the MOT problem in a data-driven way. Expand
Deep Learning for Real-Time 3D Multi-Object Detection, Localisation, and Tracking: Application to Smart Mobility
TLDR
A deep learning multi object detection and tracking technique applied to road smart mobility and an extended Kalman filter is introduced to better estimate the position of objects to improve SORT approach for 3D object tracking. Expand
Sports broadcasting and multiple object tracking with deep learning methods
Since less than a decade ago, deep learning techniques started to dominate many different fields, revolutionizing the possibilities of artificial intelligence. Seeing their potential, industrialExpand
Multiple objects tracking in the UAV system based on hierarchical deep high-resolution network
TLDR
An online Multi-Object Tracking (MOT) approach in the UAV system is proposed to handle small target detections and class imbalance challenges, which integrates the merits of deep high-resolution representation network and data association method in a unified framework. Expand
Towards Collaborative Robotics in Top View Surveillance: A Framework for Multiple Object Tracking by Detection Using Deep Learning
TLDR
A collaborative robotics framework that can assist in the detection and tracking of multiple objects in top view surveillance is presented and the generalization performance is investigated through testing the models on various sequences of top view data set. Expand
Multiple Object Trackers in OpenCV: A Benchmark
TLDR
This paper evaluates 7 trackers implemented in OpenCV against the MOT20 dataset and results are shown based on Multiple Object Tracking Accuracy (MOTA) and Multiple Object tracking Precision (MOTP) metrics. Expand
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 185 REFERENCES
Online multi-object tracking with convolutional neural networks
TLDR
A novel online multi-object tracking framework, which exploits features from multiple convolutional layers, which uses the top layer to formulate a category-level classifier and a lower layer to identify instances from one category under the intuition that lower layers contain much more details. Expand
Learning deep features for multiple object tracking by using a multi-task learning strategy
TLDR
A feature learning algorithm that simultaneously learns the common feature shared by all target objects and the individual feature of each object and can significantly improve multiple object tracking performance. Expand
Collaborative Deep Reinforcement Learning for Multi-object Tracking
TLDR
A deep prediction-decision network is developed in the C-DRL, which simultaneously detects and predicts objects under a unified network via deep reinforcement learning. Expand
Multi-Object Tracking with Correlation Filter for Autonomous Vehicle
TLDR
This paper improves the detection module by incorporating the temporal information, which is beneficial for detecting small objects, and proposes a novel compressed deep Convolutional Neural Network feature based Correlation Filter tracker for the tracking module. Expand
Instance flow based online multiple object tracking
TLDR
This work presents a method to perform online Multiple Object Tracking (MOT) of known object categories in monocular video data by exploiting state-of-the-art instance aware semantic segmentation techniques to compute 2D shape representations of target objects in each frame. Expand
Multi-target tracking using CNN-based features: CNNMTT
TLDR
Comprehensive evaluations of the algorithm (CNNMTT) reveal that the CNNMTT method achieves high quality tracking results in comparison to the state of the art while being faster and involving much less computational cost. 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
Multi-person tracking by discriminative affinity model and hierarchical association
TLDR
This paper exploits a new discriminative affinity model which combines deep convolution appearance features, random walk motion model, and object shape to robustly handle occlusion during a long period of time and proposes a hierarchical association framework. Expand
Online Multi-Target Tracking Using Recurrent Neural Networks
TLDR
This work proposes for the first time, an end-to-end learning approach for online multi-target tracking based on recurrent neural networks (RNNs) and shows promising results obtained at ~300 Hz on a standard CPU. Expand
Data Association for Multi-Object Tracking via Deep Neural Networks
TLDR
The experiment results show that the proposed network achieves considerably high recall and precision rate as the binary classifier for the assignment tasks and the performance of the tracker outperforms previous works based on DNN and comparable to other state-of-the-art methods. Expand
...
1
2
3
4
5
...