Adaptive mean-shift for automated multi object tracking

Beyan, Çiğdem; Temizel, Alptekin

doi:10.1049/iet-cvi.2011.0054

Cited by 46 publications

(28 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Here, = = 10, = 100. (2) Second, the mean-shift method [28] is used to estimate the particle position at frame +1 :…”

Section: Strategy To Track Scale Variationmentioning

confidence: 99%

“…Meanwhile, both situations have the risk to collect dirty positive and negative samples, which may degrade the online updated classifier to cause drift. The reason to introduce mean-shift method [28] into our scale tracking strategy is two unfolds. For one thing, this method computes the direction with fastest increasing probability density, a local extreme of probability density could be derived after several iterations.…”

Section: Strategy To Track Scale Variationmentioning

confidence: 99%

See 1 more Smart Citation

Real-Time Inland CCTV Ship Tracking

Xiao

2018

Mathematical Problems in Engineering

View full text Add to dashboard Cite

The predator algorithm is a representative pioneering work that achieves state-of-the-art performance on several popular visual tracking benchmarks and with great success when commercially applied to real-time face tracking in long-term unconstrained videos. However, there are two major drawbacks of predator algorithm when applied to inland CCTV (closed-circuit television) ship tracking. First, the LK short-term tracker within predator algorithm easily tends to drift if the target ship suffers partial or even full occlusion, mainly because the corner-points-like features employed by LK tracker are very sensitive to occlusion appearance change. Second, the cascaded detector within the predator algorithm searches for candidate objects in a predefined scale set, usually including 3-5 elements, which hampers the tracker to adapt to the potential diverse scale variations of the target ship. In this paper, we design a random projection based short-term tracker which can dramatically ease the tracking drift when the ship is under occlusion. Furthermore, a forward-backward feedback mechanism is proposed to estimate the scale variation between two consecutive frames. We prove that these two strategies gain significant improvements over the predator algorithm and also show that the proposed method outperforms several other state-of-the-art trackers.

show abstract

“…Here, = = 10, = 100. (2) Second, the mean-shift method [28] is used to estimate the particle position at frame +1 :…”

Section: Strategy To Track Scale Variationmentioning

confidence: 99%

Section: Strategy To Track Scale Variationmentioning

confidence: 99%

Real-Time Inland CCTV Ship Tracking

Xiao

2018

Mathematical Problems in Engineering

View full text Add to dashboard Cite

show abstract

“…The recall, precision and F-score are calculated as follows [45]: To intuitively evaluate the performance of a tracking algorithm on the whole image sequence, the average tracking error, the average recall, the average precision and average Fscore are used, which are the mean of the sum of each frame's tracking error, recall, precision and F-score, respectively.…”

Section: Page 15 Of 30mentioning

confidence: 99%

Adaptive edge-based mean shift for drastic change gray target tracking

Zheng

Mao

Chen

et al. 2015

Optik

View full text Add to dashboard Cite

“…The author shows Experimental results and proof the new method can successfully track the object under such case as merging, splitting, scale variation and scene noise. The author Bayan [13] talks about adaptive mean shift for automated multi tracking. The benefit of Gaussian mixture model is that it extracted Foreground image from video frame sequence it also eliminate the shadow and noise from video sequence.…”

Section: Section 5 Object Detection From Videomentioning

confidence: 99%