Proceedings Article10.1109/ICIEA.2019.8834374
Confidence map based KCF object tracking algorithm
Baoguo Wei,Yufei Wang,Xingjian He +2 more
- 19 Jun 2019
- pp 2187-2192
7
TL;DR: The experimental results show that the proposed approach improves success rate and precision by 7% and 8% respectively and an innovative model update mechanism to reduce the computational complexity and model contamination is proposed.
read more
Abstract: Tracking with kernelized correlation filters (KCF) is an excellent object tracking algorithm, which is widely concerned. In KCF, each candidate patch in tracked region corresponds to a confidence ratio indicating the probability of containing the target, and the patch with the maximum confidence ratio is the output. In traditional KCF, its tracking performance decreases in complex scenes and the model is liable to be contaminated due to updated every frame. To overcome these limitations, we combine all available confidence ratios to form a confidence map, then by analyzing the confidence map, we infer the tracking scene and adopt different tracking strategies. For complex scenes, we dynamically improve KCF to enhance its tracking performance. In addition, we propose an innovative model update mechanism to reduce the computational complexity and model contamination. The experimental results show that compared with the conventional KCF algorithm, the proposed approach improves success rate and precision by 7% and 8% respectively.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
IEEE Transactions on Pattern Analysis and Machine Intelligence
King-Sun Fu
- 15 Oct 2004
Abstract: Abstmct-In this correspondence, we show how to recover the motion of an observer relative to a planar surface from image brightness derivatives. We do not compute the optical flow as an intermediate step, only the spatial and temporal brightness gradients (at a minimum of eight points). We first present two iterative schemes for solving nine nonlinear equations in terms of the motion and surface parameters that are derived from a least-squares fomulation. An initial pass over the relevant image region is wed to accumulate a number of moments of the image brightness derivatives. All of the quantities used in the iteration are efficiently computed from these totals without the need to refer back to the image. We then show that either of two possible solutions can be obtained in closed form. We first solve a linear matrix equation for the elements of a 3 x 3 matrix. The eigenvalue decomposition of the symmetric part of the matrix is then used to compute the motion parameters and the plane orientation. A new compact notation allows us to show easily that there are at most two planar solutions.
2.1K
Patent
Tracking-learning-detection (TLD)-based video object tracking method
TL;DR: According to the method, the conventional TLD algorithm is improved to obtain a video object tracking algorithm which is more ideal than the T LD algorithm.
83
Towards Developing a Real-Time Hand Gesture Controlled Wheelchair
Md. Repon Islam,Md. Saiful Islam,Muhammad Sheikh Sadi +2 more
- 01 Jan 2021
TL;DR: In this paper, a control system prototype to easily navigate the wheelchair by performing finger gestures rather than operating a traditional joystick is presented. But the system is focused more on the flexibilities of wheelchair users by ensuring less movement of the hand as well as the cost of the implementation.
8
Multi-criteria Confidence Evaluation for Robust Visual Tracking
Shi Siqi,Li Nanting,Ma Yanjun,Liping Zheng +3 more
- 29 Oct 2021
TL;DR: In this paper, a multi-criteria confidence evaluation strategy is proposed to solve the challenge of visual tracking in complex environment, and three kinds of criteria are introduced to comprehensively evaluate and analyze the confidence of those tracking results obtained by adaptive spatially-regularized correlation filters (ASRCF).
1
A scale-adaptive real-time target tracking algorithm based on KCF
Fan Zhao,Kaidi Hui,Tingting Wang,Erhu Zhang,Guangfeng Lin,Yajun Chen +5 more
- 12 Jun 2020
TL;DR: This paper proposes a scale-adaptive target tracking algorithm based on KCF, which achieves a higher success rate and accuracy with less running time compared with the state-of-the- art methods.
References
Histograms of oriented gradients for human detection
Navneet Dalal,Bill Triggs +1 more
- 20 Jun 2005
TL;DR: It is shown experimentally that grids of histograms of oriented gradient (HOG) descriptors significantly outperform existing feature sets for human detection, and the influence of each stage of the computation on performance is studied.
High-Speed Tracking with Kernelized Correlation Filters
TL;DR: A new kernelized correlation filter is derived, that unlike other kernel algorithms has the exact same complexity as its linear counterpart, which is called dual correlation filter (DCF), which outperform top-ranking trackers such as Struck or TLD on a 50 videos benchmark, despite being implemented in a few lines of code.
Object Tracking Benchmark
TL;DR: An extensive evaluation of the state-of-the-art online object-tracking algorithms with various evaluation criteria is carried out to identify effective approaches for robust tracking and provide potential future research directions in this field.
Tracking-Learning-Detection
TL;DR: A novel tracking framework (TLD) that explicitly decomposes the long-term tracking task into tracking, learning, and detection, and develops a novel learning method (P-N learning) which estimates the errors by a pair of “experts”: P-expert estimates missed detections, and N-ex Expert estimates false alarms.
3.7K
Exploiting the circulant structure of tracking-by-detection with kernels
João F. Henriques,Rui Caseiro,Pedro Martins,Jorge Batista +3 more
- 07 Oct 2012
TL;DR: Using the well-established theory of Circulant matrices, this work provides a link to Fourier analysis that opens up the possibility of extremely fast learning and detection with the Fast Fourier Transform, which can be done in the dual space of kernel machines as fast as with linear classifiers.