Pose Estimation of Ad-Hoc Mobile Camera Networks

Sánta, Zsolt; Kató, Zoltán

doi:10.1109/dicta.2013.6691514

Cited by 3 publications

(14 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The aim of the proposed method is to localize the network in a three-dimensional (3D) scene with respect to a 3D structure [Sánta and Kato, 2013b]. In other words, the origin of the estimated network will be attached to a well-defined 3D location.…”

Section: The Calibration Frameworkmentioning

confidence: 99%

“…Now, let us determine the relative pose of the cameras w.r.t. a main camera [Sánta and Kato, 2013b].…”

Section: Relative Pose Of the Cameras Within The Networkmentioning

confidence: 99%

“…According to the review from Section 2.2.1, this problem can be easily solved by estimating the essential matrices acting between the main camera and the other cameras. For that purpose, ASIFT [29] features are extracted by each mobile in a distributed, parallel manner, and then the descriptors are sent to the main camera, which subsequently computes correspondences and estimates the essential matrices [Sánta and Kato, 2013b]. The necessary amount of communication depends on the size of the adopted descriptor, it is typically between 64 and 128 for ASIFT features.…”

Section: Relative Pose Of the Cameras Within The Networkmentioning

confidence: 99%

“…In our experiments, M i was at most 500 and C = 64. Each packet has to be sent only once to the main camera [Sánta and Kato, 2013b]. Of course, the size of a packet could be reduced using compression methods described in [79,80,138], but this is beyond the scope of the current work.…”

Section: Relative Pose Of the Cameras Within The Networkmentioning

confidence: 99%

“…Recall that, the main advantage of this method over other approaches is that it returns only one solution for the essential matrix. However, we have to use at least 8 point correspondences [Sánta and Kato, 2013b].…”

Section: Relative Pose Of the Cameras Within The Networkmentioning

confidence: 99%

See 4 more Smart Citations

Non-Rigid Registration of Visual Objects

Sánta

View full text Add to dashboard Cite

Sample results on the Bosphorus dataset. In each row, we show the template, the observation, the warped template and the landmark locations, respectively. In the last column, green denotes the ground truth position and red is the estimated location. While the proposed approach achieved good results near nose and mouth areas, the highest errors are near the eyebrows.. .. .. .. . 53 LIST OF FIGURES v 3.20 Comparison between the proposed, the GMMREG [5] and the CPD [6] algorithms. On the left diagram, the D GT measure is presented, while on the right we show the running times of each algorithm. The GMMREG slightly outperformed the proposed approach and the CPD algorithm in terms of alignment accuracy, but the proposed approach has the best running time.. .. .. .. .

show abstract

Section: The Calibration Frameworkmentioning

confidence: 99%

“…Now, let us determine the relative pose of the cameras w.r.t. a main camera [Sánta and Kato, 2013b].…”

Section: Relative Pose Of the Cameras Within The Networkmentioning

confidence: 99%