Video stabilization based on point feature matching technique

Abdullah, Labeeb Mohsin; Tahir, Nooritawati Md; Samad, Mustaffa

doi:10.1109/icsgrc.2012.6287181

Cited by 22 publications

(16 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Feature detection and matching stabilization method was used to remove unwanted movements in video [29,30]. The main framework of the method is based on three main stages which are motion estimation, motion smoothing, and image composition.…”

Section: A Video Stabilizing Using Features Detection Extraction Anmentioning

confidence: 99%

Preprocessing Realistic Video for Contactless Heart Rate Monitoring Using Video Magnification

Alzahrani

Whitehead

2015

2015 12th Conference on Computer and Robot Vision

View full text Add to dashboard Cite

This research seeks to improve the outcomes of Eulerian Video Magnification in real life scenarios. We address the core requirement in Eulerian Magnification that the person in the video be completely still. The proposed system preprocesses the video in multiple stages using subject targeting and stabilization. The resulting video is better suited to Eulerian Magnification restrictions. Our method enables the use of magnification in a variety of applications where motion is present such as monitoring the heart rate of a person using a treadmill. Stabilization, which is the core element of our research, was achieved through two methods. First, we used face tracking to generate a stabilized video with limited motion. Second, feature detection, extraction, and matching with skin selection were used to produce a stabilized video that is ready to be processed for measuring heart rate. However, skin tone and illumination in the environment adversely affected the results. Since heart rate is monitored by counting the subtle changes in skin redness related to blood flow, managing the skin's redness helps to produce more accurate results.

show abstract

Section: A Video Stabilizing Using Features Detection Extraction Anmentioning

confidence: 99%

Preprocessing Realistic Video for Contactless Heart Rate Monitoring Using Video Magnification

Alzahrani

Whitehead

2015

2015 12th Conference on Computer and Robot Vision

View full text Add to dashboard Cite

show abstract

“…To handle the noise they applied Kalman filtering and median filtering approaches. In [12] , the jitter of the camera is eliminated by using a feature-based approach consist of Harris Corner Detection as descriptor, Sum of Squared Differences (SSD) for finding correspondence points, and Random Sample Consensus algorithm (RANSAC) to remove the outliers.…”

Section: Video Stabilizationmentioning

confidence: 99%

Analysis to feature-based video stabilization/registration techniques within application of traffic data collection

Sadat

Viti

2015

SPIE Proceedings

View full text Add to dashboard Cite

Machine vision is rapidly gaining popularity in the field of Intelligent Transportation Systems. In particular, advantages are foreseen by the exploitation of Aerial Vehicles (AV) in delivering a superior view on traffic phenomena. However, vibration on AVs makes it difficult to extract moving objects on the ground. To partly overcome this issue, image stabilization/registration procedures are adopted to correct and stitch multiple frames taken of the same scene but from different positions, angles, or sensors. In this study, we examine the impact of multiple feature-based techniques for stabilization, and we show that SURF detector outperforms the others in terms of time efficiency and output similarity.

show abstract

“…However, most of our video data suffers from camera jitter, which can result in many false positives. Therefore, we use an image stabilization approach [19] before applying the temporal detector. Then, we analyze the pixel velocities in all fire-like regions resulting from the previous CAD step.…”

Section: Temporal Analysismentioning

confidence: 99%

Fire Detection in Videos of Violent Crowds Acquired with Handheld Devices

Moria

Albu

2015

2015 12th Conference on Computer and Robot Vision

View full text Add to dashboard Cite

Our paper proposes a novel method for fire detection in riot videos acquired with handheld cameras and smart-phones. This is a typical example of computer vision in the wild, where we have no control over the data acquisition process, and the quality of the video data varies considerably. We propose a novel spatial model for fire, based on Gaussian mixtures and on color adjacency in the visible spectrum of incandescence. We demonstrate that using this spatial model in concert with motion cues leads to highly accurate results for fire detection in noisy, complex scenes of violent crowds.

show abstract

Video stabilization based on point feature matching technique

Cited by 22 publications

References 8 publications

Preprocessing Realistic Video for Contactless Heart Rate Monitoring Using Video Magnification

Preprocessing Realistic Video for Contactless Heart Rate Monitoring Using Video Magnification

Analysis to feature-based video stabilization/registration techniques within application of traffic data collection

Fire Detection in Videos of Violent Crowds Acquired with Handheld Devices

Contact Info

Product

Resources

About