Tracking Feature Points in Uncalibrated Images with Radial Distortion

Lourenço, Miguel; Barreto, João Pedro

doi:10.1007/978-3-642-33765-9_1

Cited by 4 publications

(8 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Lourenço and Barreto show in [8] that it is possible to estimate the radial distortion in the image plane by tracking feature points between adjacent frames. Their uncalibrated KLT algorithm for images with radial distortion (uRD-KLT) starts by extracting reference templates T(x) around a set of salient points x that are detected based on image derivatives [8].…”

Section: Estimating Image Distortion At Every Frame Using Urd-kltmentioning

confidence: 99%

“…This article reports a solution for efficient and accurate focal length estimation in endoscopic video. We built on recent advances in tracking image features between frames with radial distortion [8] to show that it is possible to recover the focal length variation in sequences with zoom variation. Since we built on tracking theory, our approach is well suited for processing continuous monocular endoscopic video, does not make assumptions about camera motion [4] or scene rigidity [5], and does not require the boundary contour of the lens to be visible [7].…”

Section: Miguel Lourenço and João Barreto Want To Thank Qren-mais Cenmentioning

confidence: 99%

“…We will use the relation of Eq. 6 for recovering the focal length f at each frame by combining offline calibration of the constant parameter ξ using [1] with online estimation of η using the tracking framework of [8].…”

Section: Direct Projection Model and Single Image Calibrationmentioning

confidence: 99%

“…Their uncalibrated KLT algorithm for images with radial distortion (uRD-KLT) starts by extracting reference templates T(x) around a set of salient points x that are detected based on image derivatives [8]. Given an incoming image I(x), the goal is to align the templates T(x) with the corresponding image regions subject to the squared intensity difference and under the assumption of a 2D deformation model v(x; p), with parameters p, that accounts for both the local motion w and the global effect of distortion.…”

Section: Estimating Image Distortion At Every Frame Using Urd-kltmentioning

confidence: 99%

See 3 more Smart Citations

Continuous Zoom Calibration by Tracking Salient Points in Endoscopic Video

Lourenço

Barreto

Fonseca³

et al. 2014

Medical Image Computing and Computer-Assisted Intervention – MICCAI 2014

Self Cite

View full text Add to dashboard Cite

Abstract. Many image-based systems for aiding the surgeon during minimally invasive surgery require the endoscopic camera to be calibrated at all times. This article proposes a method for accomplishing this goal whenever the camera has optical zoom and the focal length changes during the procedure. Our solution for online calibration builds on recent developments in tracking salient points using differential image alignment, is well suited for continuous operation, and makes no assumptions about the camera motion or scene rigidity. Experimental validation using both a phantom model and in vivo data shows that the method enables accurate estimation of focal length when the zoom varies, avoiding the need to explicitly recalibrate during surgery. To the best of our knowledge this the first work proposing a practical solution for online zoom calibration in the operation room.

show abstract

Section: Estimating Image Distortion At Every Frame Using Urd-kltmentioning

confidence: 99%

Section: Miguel Lourenço and João Barreto Want To Thank Qren-mais Cenmentioning

confidence: 99%

Section: Direct Projection Model and Single Image Calibrationmentioning

confidence: 99%

Section: Estimating Image Distortion At Every Frame Using Urd-kltmentioning

confidence: 99%

See 2 more Smart Citations

Continuous Zoom Calibration by Tracking Salient Points in Endoscopic Video

Lourenço

Barreto

Fonseca³

et al. 2014

Medical Image Computing and Computer-Assisted Intervention – MICCAI 2014

Self Cite

View full text Add to dashboard Cite

show abstract

“…While it is very common to simply use Brownian or constantvelocity motion models, with parameters set a priori, their generality hurts them in many specific situations. Lourenço and Barreto [19] show the importance of an appropriate motion model in the case of scenes with radial distortion. If tracking an object with a known nature, a physical model can better dictate the system's predictions [5].…”

Section: Related Workmentioning

confidence: 99%

Motion Models that Only Work Sometimes

Cifuentes¹,

Sturzel²,

Jurie³

et al. 2012

Procedings of the British Machine Vision Conference 2012

View full text Add to dashboard Cite

It is too often that tracking algorithms lose track of interest points in image sequences. This persistent problem is difficult because the pixels around an interest point change in appearance or move in unpredictable ways. In this paper we explore how classifying videos into categories of camera motion improves the tracking of interest points, by selecting the right specialist motion model for each video. As a proof of concept, we enumerate a small set of simple categories of camera motion and implement their corresponding specialized motion models. We evaluate the strategy of predicting the most appropriate motion model for each test sequence. Within the framework of a standard Bayesian tracking formulation, we compare this strategy to two standard motion models. Our tests on challenging real-world sequences show a significant improvement in tracking robustness, achieved with different kinds of supervision at training time.frame 0 frame 10 frame 20 Figure 1: Example sequence with overlaid box showing the output of our specialized "forward" motion model, where the velocity and scale of objects approaching the camera tend to increase. Neither Brownian nor constant-velocity motion models are as successful at tracking interest points here.

show abstract

CoMaL: Good Features to Match on Object Boundaries

Ravindran

Mittal

2016

2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

View full text Add to dashboard Cite

Tracking Feature Points in Uncalibrated Images with Radial Distortion

Cited by 4 publications

References 14 publications

Continuous Zoom Calibration by Tracking Salient Points in Endoscopic Video

Continuous Zoom Calibration by Tracking Salient Points in Endoscopic Video

Motion Models that Only Work Sometimes

CoMaL: Good Features to Match on Object Boundaries

Contact Info

Product

Resources

About