Related references
Note: Only part of the references are listed.
Article
Computer Science, Artificial Intelligence
Pengfei Zhu et al.
Summary: Drones equipped with cameras have been widely used in various fields, and the automatic understanding of visual data collected from drones has become highly demanding. To promote the development of object detection and tracking algorithms, challenge workshops have been organized, and a large-scale drone captured dataset, VisDrone, has been provided. The dataset enables extensive evaluation and investigation of visual analysis algorithms. The paper analyzes the current state of the field and proposes future directions.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Yifu Zhang et al.
Summary: This method improves the performance of multi-object tracking by associating almost every detection box, effectively solving the problem of true object missing and fragmented trajectories caused by low score detection boxes being discarded. Applied to multiple trackers, this method consistently achieves improvement on IDF1 score.
COMPUTER VISION, ECCV 2022, PT XXII
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Peize Sun et al.
Summary: This article introduces a large-scale dataset for multi-human tracking, named DanceTrack, which primarily consists of group dancing videos. The aim is to encourage the development of MOT algorithms that rely on motion analysis rather than visual discrimination. The article highlights the biases in existing tracking datasets and evaluates the performance of several state-of-the-art trackers on DanceTrack.
2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2022)
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Tim Meinhardt et al.
Summary: This study proposes an end-to-end trainable multi-object tracking approach called TrackFormer, based on an encoder-decoder Transformer architecture. TrackFormer achieves outstanding performance in track initialization, identity, and spatio-temporal trajectory reasoning, and introduces the attention mechanism. Through self- and encoder-decoder attention on global frame-level features, additional graph optimization or modeling of motion and/or appearance is omitted.
2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)
(2022)
Article
Computer Science, Artificial Intelligence
Jonathon Luiten et al.
Summary: The higher order tracking accuracy (HOTA) is a novel evaluation metric for multi-object tracking that balances accurate detection, association, and localization. It decomposes into sub-metrics to evaluate different error types separately, providing clear analysis of tracking performance. The HOTA scores align better with human visual evaluation of tracking performance compared to established metrics.
INTERNATIONAL JOURNAL OF COMPUTER VISION
(2021)
Article
Computer Science, Artificial Intelligence
Patrick Dendorfer et al.
Summary: Standardized benchmarks are crucial for pushing the performance of computer vision algorithms, with leaderboards offering objective measures of performance. MOTChallenge is a benchmark for single-camera Multiple Object Tracking launched in late 2014, focusing on multiple people tracking and aiming to create a framework for standardized evaluation of tracking methods.
INTERNATIONAL JOURNAL OF COMPUTER VISION
(2021)
Review
Computer Science, Artificial Intelligence
Wenhan Luo et al.
Summary: This review comprehensively examines the problem of Multiple Object Tracking (MOT) and proposes interesting directions for future research. By analyzing existing methods and experimental results, some fundamental agreements in the field have been verified.
ARTIFICIAL INTELLIGENCE
(2021)
Article
Computer Science, Artificial Intelligence
Yifu Zhang et al.
Summary: Multi-object tracking is a crucial problem in computer vision, and formulating it as multi-task learning of object detection and re-ID in a single network can lead to joint optimization of the two tasks. However, competition between the tasks needs to be addressed, and the proposed FairMOT method based on CenterNet architecture achieves high accuracy for both detection and tracking through detailed designs and empirical studies.
INTERNATIONAL JOURNAL OF COMPUTER VISION
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Hexin Bai et al.
Summary: This paper introduces contributions to the study of Generic Multiple Object Tracking (GMOT), including the construction of the GMOT-40 dataset, design of baseline algorithms, and evaluations. GMOT is expected to receive more attention in future research.
2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Peng Dai et al.
Summary: This paper proposes a novel proposal-based learnable framework for multiple object tracking, which models the process into different stages and achieves significant performance improvement.
2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Song Guo et al.
Summary: The paper introduces a novel unified model that achieves synergy between position prediction and embedding association through temporal-aware target attention and identity-aware memory aggregation. The new model outperforms existing methods in MOT field experiments.
2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Bing Shuai et al.
Summary: This paper introduces a region-based Siamese Multi-Object Tracking network called SiamMOT to improve online multi-object tracking by modeling motion. Experimental results show that SiamMOT performs superiorly in MOT and outperforms the winners of ACM MM'20 HiEve Grand Challenge on the HiEve dataset.
2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Alexander Mathis et al.
Summary: This study investigates the generalization ability of neural networks for pose estimation and introduces a new dataset of 30 horses for testing model robustness in both within- and out-of-domain scenarios. The findings suggest that architectures pretrained on ImageNet perform better on both within- and out-of-domain data, and transfer learning is beneficial for out-of-domain robustness in pose estimation tasks.
2021 IEEE WINTER CONFERENCE ON APPLICATIONS OF COMPUTER VISION (WACV 2021)
(2021)
Article
Computer Science, Artificial Intelligence
Gioele Ciaparrone et al.
Article
Computer Science, Artificial Intelligence
Longyin Wen et al.
COMPUTER VISION AND IMAGE UNDERSTANDING
(2020)
Article
Computer Science, Theory & Methods
Patrick Emami et al.
ACM COMPUTING SURVEYS
(2020)
Article
Multidisciplinary Sciences
Praneet C. Bala et al.
NATURE COMMUNICATIONS
(2020)
Proceedings Paper
Engineering, Electrical & Electronic
Peng Chu et al.
2019 IEEE WINTER CONFERENCE ON APPLICATIONS OF COMPUTER VISION (WACV)
(2019)
Proceedings Paper
Computer Science, Artificial Intelligence
Jinkun Cao et al.
2019 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2019)
(2019)
Article
Computer Science, Artificial Intelligence
Olga Russakovsky et al.
INTERNATIONAL JOURNAL OF COMPUTER VISION
(2015)
Proceedings Paper
Computer Science, Artificial Intelligence
Yumi Iwashita et al.
2014 22ND INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR)
(2014)
Article
Engineering, Electrical & Electronic
Keni Bernardin et al.
EURASIP JOURNAL ON IMAGE AND VIDEO PROCESSING
(2008)