Dataset Information

Robust Video Stabilization Using Particle Keypoint Update and l?-Optimized Camera Path.

ABSTRACT: Acquisition of stabilized video is an important issue for various type of digital cameras. This paper presents an adaptive camera path estimation method using robust feature detection to remove shaky artifacts in a video. The proposed algorithm consists of three steps: (i) robust feature detection using particle keypoints between adjacent frames; (ii) camera path estimation and smoothing; and (iii) rendering to reconstruct a stabilized video. As a result, the proposed algorithm can estimate the optimal homography by redefining important feature points in the flat region using particle keypoints. In addition, stabilized frames with less holes can be generated from the optimal, adaptive camera path that minimizes a temporal total variation (TV). The proposed video stabilization method is suitable for enhancing the visual quality for various portable cameras and can be applied to robot vision, driving assistant systems, and visual surveillance systems.

SUBMITTER: Jeon S

PROVIDER: S-EPMC5336040 | biostudies-literature | 2017 Feb

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Robust Video Stabilization Using Particle Keypoint Update and l₁-Optimized Camera Path.

Jeon Semi S Yoon Inhye I Jang Jinbeum J Yang Seungji S Kim Jisung J Paik Joonki J

Sensors (Basel, Switzerland) 20170210 2

Acquisition of stabilized video is an important issue for various type of digital cameras. This paper presents an adaptive camera path estimation method using robust feature detection to remove shaky artifacts in a video. The proposed algorithm consists of three steps: (i) robust feature detection using particle keypoints between adjacent frames; (ii) camera path estimation and smoothing; and (iii) rendering to reconstruct a stabilized video. As a result, the proposed algorithm can estimate the ...[more]

PMID: 28208622

Similar Datasets

Project description:Background:Optical mapping of transmembrane voltage and intracellular calcium is a powerful tool for investigating cardiac physiology and pathophysiology. However, simultaneous dual mapping of two fluorescent probes remains technically challenging. We introduce a novel, easy-to-use approach that requires a path splitter, single camera and excitation light to simultaneously acquire voltage and calcium signals from whole heart preparations, which can be applied to other physiological models - including neurons and isolated cardiomyocytes. Results:Complementary probes were selected that could be excited with a single wavelength light source. Langendorff-perfused hearts (rat, swine) were stained and imaged using a sCMOS camera outfitted with an optical path splitter to simultaneously acquire two emission fields at high spatial and temporal resolution. Voltage (RH237) and calcium (Rhod2) signals were acquired concurrently on a single sensor, resulting in two 384 × 256 images at 814 frames per second. At this frame rate, the signal-to-noise ratio was 47 (RH237) and 85 (Rhod2). Imaging experiments were performed on small rodent hearts, as well as larger pig hearts with sufficient optical signals. In separate experiments, each dye was used independently to assess crosstalk and demonstrate signal specificity. Additionally, the effect of ryanodine on myocardial calcium transients was validated - with no measurable effect on the amplitude of optical action potentials. To demonstrate spatial resolution, ventricular tachycardia was induced - resulting in the novel finding that spatially discordant calcium alternans can be present in different regions of the heart, even when electrical alternans remain concordant. The described system excels in providing a wide field of view and high spatiotemporal resolution for a variety of cardiac preparations. Conclusions:We report the first multiparametric mapping system that simultaneously acquires calcium and voltage signals from cardiac preparations, using a path splitter, single camera and excitation light. This approach eliminates the need for multiple cameras, excitation light patterning or frame interleaving. These features can aid in the adoption of dual mapping technology by the broader cardiovascular research community, and decrease the barrier of entry into panoramic heart imaging, as it reduces the number of required cameras.

Dataset Information

Robust Video Stabilization Using Particle Keypoint Update and l?-Optimized Camera Path.

Publications

Robust Video Stabilization Using Particle Keypoint Update and l₁-Optimized Camera Path.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets