A Curvature Based Method for Blind Mesh Visual Quality Assessment Using a General Regression Neural Network
Résumé
No-reference quality assessment is a challenging issue due to the non-existence of any information related to the reference and the unknown distortion type. The main goal is to design a computational method to objectively predict the human perceived quality of a distorted mesh and deal with the practical situation when the reference is not available. In this work, we design a no reference method that relies on the general regression neural network (GRNN). Our network is trained using the mean curvature which is an important perceptual feature representing the visual aspect of a 3D mesh. Relatively to the human subjective scores, the trained network successfully assesses the visual quality, in addition, the experimental results show that the proposed method provides good correlations with the subject scores and competitive scores comparing to some influential and effective full and reduced reference existing metrics.
Mots clés
computational geometry
learning (artificial intelligence)
mesh generation
neural nets
regression analysis INSPEC: Non-Controlled Indexing curvature based method
blind mesh visual quality assessment
no-reference quality assessment
computational method
human perceived quality prediction
distorted mesh
GRNN
mean curvature
perceptual feature
visual aspect representation
3D mesh
human subjective scores
general regression neural network training
visual quality
subject scores
competitive scores Author Keywords Blind mesh visual quality assessment
general regression neural network
mean opinion scores
predicted objective scores
feature learning