Occlusion analysis: Learning and utilising depth maps in object tracking

Greenhill, D.; Renno, J.; Orwell, James; Jones, Graeme A.

doi:10.1016/j.imavis.2006.12.007

Cited by 30 publications

(17 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our localization and RBPF-based tracking algorithm is factorized into the goal set distribution, object set distribution, robot distribution, and the last state set distribution at time k − 1 in (16). Object tracking is similar to (12) but it is conditioned on the robot position where the uncertainty of the robot localization is taken into account,…”

Section: Localization and Pomot Algorithmmentioning

confidence: 99%

“…The information of the local color, texture, and spatial features relative to the centers of objects assists the online sampling and position estimation [15]. The occlusion problem can be also solved with the aid of depth maps [16]. However, such image processing techniques cannot be applied to the laser range finder data since there is neither 2D foreground information or the partially unoccluded object information available.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Self-Localization and Stream Field Based Partially Observable Moving Object Tracking

Tseng

Tang

2009

EURASIP J. Adv. Signal Process.

View full text Add to dashboard Cite

Recommended by Fredrik GustafssonSelf-localization and object tracking are key technologies for human-robot interactions. Most previous tracking algorithms focus on how to correctly estimate the position, velocity, and acceleration of a moving object based on the prior state and sensor information. What has been rarely studied so far is how a robot can successfully track the partially observable moving object with laser range finders if there is no preanalysis of object trajectories. In this case, traditional tracking algorithms may lead to the divergent estimation. Therefore, this paper presents a novel laser range finder based partially observable moving object tracking and self-localization algorithm for interactive robot applications. Dissimilar to the previous work, we adopt a stream field-based motion model and combine it with the Rao-Blackwellised particle filter (RBPF) to predict the object goal directly. This algorithm can keep predicting the object position by inferring the interactive force between the object goal and environmental features when the moving object is unobservable. Our experimental results show that the robot with the proposed algorithm can localize itself and track the frequently occluded object. Compared with the traditional Kalman filter and particle filter-based algorithms, the proposed one significantly improves the tracking accuracy.

show abstract

Section: Localization and Pomot Algorithmmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Self-Localization and Stream Field Based Partially Observable Moving Object Tracking

Tseng

Tang

2009

EURASIP J. Adv. Signal Process.

View full text Add to dashboard Cite

show abstract

“…To deal with static occlusion, some researchers establish a depth map of the walls, entrances, and other barriers in the tracking scene before tracking and change the template by computing their positional relationship to the target [4] . Depth marking can significantly reduce the problem of static occlusion: the fitting probability is more likely to be 0.5, while the algorithm without occlusion analysis drops significantly below 0.05.…”

Section: Introductionmentioning

confidence: 99%

Optimization of the occlusion strategy in visual tracking

Wen

Zhang

2016

Tinshhua Sci. Technol.

View full text Add to dashboard Cite

Interference and anti-interference are two opposite and important issues in visual tracking. Occlusion interference can disguise the features of a target and can also be used as an effective benchmark to determine whether a tracking algorithm is reliable. In this paper, we proposed an inner Particle Swarm Optimization (PSO) algorithm to locate the optimal occlusion strategy under different tracking conditions and to identify the most effective occlusion positions and direction of movement to allow a target to evade tracking. This algorithm improved the standard PSO process in three ways. First, it introduced a death process, which greatly reduced the time cost of optimization. Second, it used statistical data to determine the fitness value of the particles so that the fitness more accurately described the tracking. Third, the algorithm could avoid being trapped in local optima, as the fitness changes with time. Experimental results showed that this algorithm was able to identify a global optimal occlusion strategy that can disturb the tracking machine with 86.8% probability over more than 10 000 tracking processes. In addition, it reduced the time cost by approximately 80%, compared with conventional PSO algorithms.

show abstract

“…A visual approach using video cameras is employed in most applications, when objects localization and tracking is based either on color hystograms [1][2][3], illumination changes [4], occlusion [5][6][7], appearance [7,8] or scale variations [9]. Infrared techniques can also be applied [10][11][12].…”

Section: Introductionmentioning

confidence: 99%

3D Localization and Tracking of Objects Using Miniature Microphones

Ionescu¹,

Carotenuto

Urbani³

2011

WSN

View full text Add to dashboard Cite

A system for accurate localization and tracking of remote objects is introduced, which employs a reference frame of four coplanar ultrasound sources as transmitters and miniature microphones that equip the remote objects as receivers. The transmitters are forced to emit pulses in the 17 -40 kHz band. A central processing unit, knowing the positions of the transmitters and the time of flight of the ultrasound signals until they reach the microphones, computes the positions of the microphones, identifying and discarding possible false signals due to echoes and environmental noise. Once the microphones are localized, the position of the object is computed by finding the placement of the geometrical reconstructed object that fits best with the calculated microphones positions. The operating principle of the localization system is based on successive frames. The data are processed in parallel for all the microphones that equip the remote objects, leading to a high repetition rate of localization frames. In the proposed prototype, all the computation, including signal filtering, time of flight detection, localization and results display, is carried out about 25 times per second on a notebook PC.

show abstract

Occlusion analysis: Learning and utilising depth maps in object tracking

Cited by 30 publications

References 20 publications

Self-Localization and Stream Field Based Partially Observable Moving Object Tracking

Self-Localization and Stream Field Based Partially Observable Moving Object Tracking

Optimization of the occlusion strategy in visual tracking

3D Localization and Tracking of Objects Using Miniature Microphones

Contact Info

Product

Resources

About