An Optical Flow-Based Full Reference Video Quality Assessment Algorithm


We tend to present a easy yet effective optical flow-primarily based full-reference video quality assessment (FR-VQA) algorithm for assessing the perceptual quality of natural videos. Our algorithm relies on the premise that native optical flow statistics are full of distortions and also the deviation from pristine flow statistics is proportional to the quantity of distortion. We tend to characterize the native flow statistics using the mean, the standard deviation, the coefficient of variation (CV), and also the minimum eigenvalue ( ) of the native flow patches. Temporal distortion is estimated as the change in the CV of the distorted flow with respect to the reference flow, and the correlation between of the reference and of the distorted patches. We rely on the sturdy multi-scale structural similarity index for spatial quality estimation. The computed temporal and spatial distortions, therefore, are then pooled employing a perceptually motivated heuristic to come up with a spatio-temporal quality score. The proposed technique is shown to be competitive with the state-of-the-art when evaluated on the LIVE SD database, the EPFL Polimi SD database, and also the LIVE Mobile HD database. The distortions considered in these databases embody those thanks to compression, packet-loss, wireless channel errors, and rate-adaptation. Our algorithm is versatile enough to allow for any sturdy FR spatial distortion metric for spatial distortion estimation. Further, the proposed technique isn't solely parameter-free but conjointly freelance of the choice of the optical flow algorithm. Finally, we tend to show that the replacement of the optical flow vectors in our proposed technique with the a lot of coarser block motion vectors additionally leads to an acceptable FR-VQA algorithm. Our algorithm is named the flow similarity index.

Did you like this research project?

To get this research project Guidelines, Training and Code... Click Here

PROJECT TITLE : Deep Neural Network Regression for Automated Retinal Layer Segmentation in Optical Coherence Tomography Images ABSTRACT: The quantification of layer information in early diagnosis of retinal disorders, the primary
PROJECT TITLE :Optimum Monthly Based Selection of Ground Stations for Optical Satellite Networks - 2018ABSTRACT:Satellite communication networks at optical frequencies are proposed to produce a terribly high throughput for backhauling
PROJECT TITLE :Energy-Efficient Transponder Configuration for FMF-Based Elastic Optical Networks - 2018ABSTRACT:We propose an energy-efficient procedure for transponder configuration in few-mode fiber-based elastic optical networks
PROJECT TITLE :Capacity Bounds and High-SNR Capacity of MIMO Intensity-Modulation Optical Channels - 2018ABSTRACT:The capability of the intensity modulation direct detection multiple-input-multiple-output channel is studied. Therein,
PROJECT TITLE : Depth Reconstruction From Sparse Samples: Representation, Algorithm, and Sampling - 2015 ABSTRACT: The fast development of 3D technology and computer vision applications has motivated a thrust of methodologies

Ready to Complete Your Academic MTech Project Work In Affordable Price ?

Project Enquiry