Mean-shift blob tracking through scale space

Collins, Robert T.

doi:10.1109/cvpr.2003.1211475

Cited by 608 publications

(406 citation statements)

References 8 publications

Supporting

Mentioning

381

Contrasting

Unclassified

Order By: Relevance

“…In addition, histogram-based representations are not designed to handle scale change although some efforts have been made to address this problem. 28,29 Holistic appearance models based on raw intensity values are used in the Kanade-Lucas-Tomasi algorithm, 30 the incremental subspace learning tracking method, 1 the incremental tensor subspace learning method 31 and the 1 -minimization based tracker. 7 However, tracking methods based on holistic representation are sensitive to partial occlusion and motion blur.…”

Section: Object Representationmentioning

confidence: 99%

An experimental comparison of online object-tracking algorithms

et al. 2011

View full text Add to dashboard Cite

This paper reviews and evaluates several state-of-the-art online object tracking algorithms. Notwithstanding decades of efforts, object tracking remains a challenging problem due to factors such as illumination, pose, scale, deformation, motion blur, noise, and occlusion. To account for appearance change, most recent tracking algorithms focus on robust object representations and effective state prediction. In this paper, we analyze the components of each tracking method and identify their key roles in dealing with specific challenges, thereby shedding light on how to choose and design algorithms for different situations. We compare state-of-the-art online tracking methods including the IVT, 1 VRT, 2 FragT, 3 BoostT, 4 SemiT, 5 BeSemiT, 6 L1T, 7 MILT, 8 VTD 9 and TLD 10 algorithms on numerous challenging sequences, and evaluate them with different performance metrics. The qualitative and quantitative comparative results demonstrate the strength and weakness of these algorithms.

show abstract

Section: Object Representationmentioning

confidence: 99%

An experimental comparison of online object-tracking algorithms

et al. 2011

View full text Add to dashboard Cite

show abstract

“…Here, we review the basic concepts of kernel-based tracking using the terminology and notation similar to [3,11].…”

Section: Mean Shift Tracking Of Locationmentioning

confidence: 99%

“…In [3], the location estimation problem is solved by optimizing the sample estimate of the Bhattacharyya coefficient:ρ…”

Section: Mean Shift Tracking Of Locationmentioning

confidence: 99%

“…Thus, we compare results from three implementations, namely mean shift, single kernel-based SSD and the multiple kernel-based SSD. The kernel used in the single kernelbased and the mean shift approach is the Epanechnikov kernel used by [3] for mean shift tracking. For the multi-kernel approach, we use the triangular kernel discussed in Section 4.…”

Section: Tracking Locationmentioning

confidence: 99%

“…Recently, these methods have been introduced to solve visual tracking problems involving location [3] and location and scale [8]. These techniques track a target region that is described as a spatially-weighted intensity histogram.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations