The Sixth Visual Object Tracking VOT2018 Challenge Results

Kristan, Matej; Leonardis, Aleš; Matas, Jiřı́; Felsberg, Michael; Pflugfelder, Roman; Zajc, Luka Čehovin; Vojíř, Tomáš; Bhat, Goutam; Lukežič, Alan; Eldesokey, Abdelrahman; Fernández, Gustavo J.; García-Martín, Álvaro; Iglesias-Arias, Álvaro; Alatan, A. Aydın; González-García, Abel; Petrosino, Alfredo; Memarmoghadam, Alireza; Vedaldi, Andrea; Muhič, Andrej; He, Anfeng; Smeulders, A.W.M.; Perera, Asanka G.; Li, Bo; Chen, Boyu; Kim, Changick; Xu, Changsheng; Xiong, Chao; Tian, Cuihua; Luo, Chong; Sun, Chong; Hao, Cong; Kim, Daijin; Mishra, Deepak; Chen, Deming; Dong, Wei; Wee, Dongyoon; Gavves, Efstratios; Gundogdu, Erhan; Velasco-Salido, Erik; Khan, Fahad Shahbaz; Yang, Fan; Zhao, Fei; Li, Feng; Battistone, Francesco; Ath, George De; Subrahmanyam, Gorthi R. K. Sai; Bastos, Guilherme Sousa; Ling, Haibin; Galoogahi, Hamed Kiani; Lee, Hankyeol; Li, Haojie; Zhao, Haojie; Fan, Heng; Zhang, Honggang; Possegger, Horst; Li, Houqiang; Lu, Huchuan; Huang, Zhi; Li, Huiyun; Lee, Hyemin; Chang, Hyung Jin; Drummond, Isabela; Valmadre, Jack; Martin, Jaime Spencer; Chahl, Javaan; Choi, Jin Young; Li, Jing; Wang, Jinqiao; Qi, Jinqing; Sung, Jinyoung; Johnander, Joakim; Henriques, João F.; Choi, Jongwon; Weijer, Joost van de; Herranz, Jorge Rodríguez; Martínez, José María Álvarez; Kittler, Josef; Zhuang, Junfei; Gao, Junyu; Grm, Klemen; Zhang, Lichao; Wang, Lijun; Yang, Lei; Rout, Litu; Liu, Si; Bertinetto, Luca; Chu, Lutao; Che, Manqiang; Maresca, Mario Edoardo; Danelljan, Martin; Yang, Ming–Hsuan; Abdelpakey, Mohamed H.; Shehata, Mohamed; Kang, Myounggu; Lee, Namhoon; Wang, Ning; Mikšík, Ondřej; Moallem, Payman; Vicente-Moñivar, Pablo; Senna, Pedro; Li, Peixia; Torr, Philip H. S.; Raju, Priya Mariam; Qian, Ruihe; Wang, Qiang; Zhou, Qin; Guo, Qing; Martín-Nieto, Rafael; Tao, Ran; Bowden, Richard; Everson, Richard; Wang, Run‐Ling; Yun, Sangdoo; Choi, Seokeon; Vivas, Sergio; Bai, Song; Huang, Shuangping; Wu, Sihang; Hadfield, Simon; Wang, Siwen; Golodetz, Stuart; Tang, Ming; Xu, Tianyang; Zhang, Tianzhu; Fischer, Tobias; Santopietro, Vincenzo; Štruc, Vitomir; Wang, Wei; Zuo, Wangmeng; Feng, Wei; Wu, Wei; Zou, Wei; Hu, Weiming; Zhou, Wengang; Zeng, Wenjun; Zhang, Xiaofan; Wu, Xiaohe; Wu, Xiao; Tian, Xinmei; Li, Yan; Lu, Yan; Law, Yee Wei; Wu, Yi; Demiris, Yiannis; Yang, Yicai; Jiao, Yifan; Li, Yuhong; Zhang, Yunhua; Sun, Yuxuan; Zhang, Zheng; Zhu, Zheng; Feng, Zhen; Wang, Zhihui; He, Zhiqun

doi:10.1007/978-3-030-11009-3_1

Cited by 438 publications

(669 citation statements)

References 82 publications

Supporting

Mentioning

668

Contrasting

Unclassified

Order By: Relevance

“…The proposed method, DSP, reaches the real-time (20 fps [43]) tracker computation, which is 21.73 for the standard fast motion. It is faster due to the small number of FFTs called.…”

Section: Dsp Compared To State-of-theart Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Dynamic swarm particle for fast motion vehicle tracking

2019

View full text Add to dashboard Cite

Nowadays, the broad availability of cameras and embedded systems makes the application of computer vision very promising as a supporting technology for intelligent transportation systems, particularly in the field of vehicle tracking. Although there are several existing trackers, the limitation of using low‐cost cameras, besides the relatively low processing power in embedded systems, makes most of these trackers useless. For the tracker to work under those conditions, the video frame rate must be reduced to decrease the burden on computation. However, doing this will make the vehicle seem to move faster on the observer's side. This phenomenon is called the fast motion challenge. This paper proposes a tracker called dynamic swarm particle (DSP), which solves the challenge. The term particle refers to the particle filter, while the term swarm refers to particle swarm optimization (PSO). The fundamental concept of our method is to exploit the continuity of vehicle dynamic motions by creating dynamic models based on PSO. Based on the experiments, DSP achieves a precision of 0.896 and success rate of 0.755. These results are better than those obtained by several other benchmark trackers.

show abstract

“…The proposed method, DSP, reaches the real-time (20 fps [43]) tracker computation, which is 21.73 for the standard fast motion. It is faster due to the small number of FFTs called.…”

Section: Dsp Compared To State-of-theart Methodsmentioning

confidence: 99%

“…This affects the precision and success rate performance. The proposed method, DSP, reaches the real-time (20 fps [43]) tracker computation, which is 21.73 for the standard fast motion. For medium and extreme fast motions, DSP needs more particle per iteration so the computation is slightly slower.…”

Section: Dsp Compared To State-of-theart Methodsmentioning

confidence: 99%

Dynamic swarm particle for fast motion vehicle tracking

2019

View full text Add to dashboard Cite

show abstract

“…Apart from the improvements achieved by sophisticated mathematical formulations, advanced DCF trackers tend to employ powerful deep CNN features for boosting the performance [13,14,15]. The trackers equipped with robust deep CNN features have outperformed those with traditional hand-crafted features in recent VOT challenges [5], while feature selection has been demonstrated as one of the most essential mechanisms enabling improved tracking performance [16,17]. Despite the recent success with graphics processing unit (GPU) implementations, it is time-consuming to extract deep CNN features and to learn complex appearance models involving a high-volume of variables in the DCF formulation on-line.…”

Section: Introductionmentioning

confidence: 99%

An accelerated correlation filter tracker

Xu¹,

Feng²,

Wu³

et al. 2020

Pattern Recognition

Self Cite

View full text Add to dashboard Cite

Recent visual object tracking methods have witnessed a continuous improvement in the state-of-the-art with the development of efficient discriminative correlation filters (DCF) and robust deep neural network features. Despite the outstanding performance achieved by the above combination, existing advanced trackers suffer from the burden of high computational complexity of the deep feature extraction and online model learning. We propose an accelerated ADMM optimisation method obtained by adding a momentum to the optimisation sequence iterates, and by relaxing the impact of the error between DCF parameters and their norm. The proposed optimisation method is applied to an innovative formulation of the DCF design, which seeks the most discriminative spatially regularised feature channels. A further speed up is achieved by an adaptive initialisation of the filter optimisation process. The significantly increased convergence of the DCF filter is demonstrated by establishing the optimisation process equivalence with a continuous dynamical system for which the convergence properties can readily be derived. The experimental results obtained on several well-known benchmarking datasets demonstrate the efficiency and robustness of the proposed ACFT method, with a tracking accuracy comparable to the start-of-the-art trackers.

show abstract

“…The field of visual object tracking has received huge attention in recent years [20,6,7]. The developed techniques cover many problems and various methods were proposed, such as single object tracking [9,1,19,17], long-term tracking [10], methods with redetection and learning [3,13,12,18], or multi-view [8] and multi-camera [14] methods.…”

Section: Introductionmentioning

confidence: 99%

Non-causal Tracking by Deblatting

Rozumnyi

Kotera

Šroubek

et al. 2019

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

0000−0001−9874−1349] , Jan Kotera 2[0000−0001−5528−5531] , FilipŠroubek 2[0000−0001−6835−4911] , and Jiří Matas 1[0000−0003−0863−4844]Abstract. Tracking by Deblatting 1 stands for solving an inverse problem of deblurring and image matting for tracking motion-blurred objects. We propose noncausal Tracking by Deblatting which estimates continuous, complete and accurate object trajectories. Energy minimization by dynamic programming is used to detect abrupt changes of motion, called bounces. High-order polynomials are fitted to segments, which are parts of the trajectory separated by bounces. The output is a continuous trajectory function which assigns location for every realvalued time stamp from zero to the number of frames. Additionally, we show that from the trajectory function precise physical calculations are possible, such as radius, gravity or sub-frame object velocity. Velocity estimation is compared to the high-speed camera measurements and radars. Results show high performance of the proposed method in terms of Trajectory-IoU, recall and velocity estimation. Recently, a method called Tracking by Deblatting 1 (TbD) has been introduced by Kotera et al.[5] to alleviate some of these restrictions. TbD performs significantly better than [15] and for a larger range of scenarios. The method solves two inverse problems of deblurring and image matting, and estimates object trajectories as piece-wise parabolic curves in each frame individually.In its core, TbD assumes causal processing of video frames, i.e. the trajectory reported in the current frame is estimated using only information from previous frames.

show abstract

The Sixth Visual Object Tracking VOT2018 Challenge Results

Cited by 438 publications

References 82 publications

Dynamic swarm particle for fast motion vehicle tracking

Dynamic swarm particle for fast motion vehicle tracking

An accelerated correlation filter tracker

Non-causal Tracking by Deblatting

Contact Info

Product

Resources

About