Facial Paralysis Dataset

Citation Author(s):
Nazil
Perveen
Research Scholar
Submitted by:
Nazil Perveen
Last updated:
Tue, 08/27/2019 - 02:40
DOI:
10.21227/6dsz-7d76
Data Format:
License:
3597 Views
Categories:
Keywords:
1
3 ratings - Please login to submit your rating.

Abstract 

Facial paralysis is the loss of facial muscle movement

either in one side or both sides of the face due to the facial

nerve damage. Currently, the subjective assessments are widely

used techniques to determine the measure of degree with which

the patient is affected. However, the subjective assessments are

highly dependant on the expert’s view and a few sets of grading

rules. In this paper, the quantitative assessment to measure the

degree of facial paralysis is proposed. The video database of

facially paralyzed patients, which consists of seven different views

and multiple subjects with ten different expressions are collected

under three experts supervision. Inorder to capture the variations

present in multiple views and subjects across all the expressions, a

large Gaussian mixture model (GMM) is trained. A feature vector

is obtained from each expression using a maximum a posteriori

adaptation (MAP). The dimension of the adapted feature vector

is very high and contains redundant attributes. So, we reduce

the dimension of the feature vector using factor analysis, which

contains pertinent attributes. The proposed work is evaluated on

the video database of 39 facially paralyzed patients of different

age groups and gender. Based on the facial paralysis effect,

experts assign subjective scores to patients using Yanagihara

grading rules, which are further used as ground truth. We also

show the efficacy of the proposed approach by measuring the

different degree of facial paralysis for all 10 types of expressions

better than existing approaches for quantitative assessment.

Instructions: 
The facial paralysis dataset consists of train_gz files folder and test_gz file folders in gz format, where improved dense trajectories features: (http://lear.inrialpes.fr/~wang/improved_trajectories) are present, you can use these features for further training and evaluation. We will upload soon other features like 3Dconvolution features and STIP features for more reproducibility of the dataset. Due to privacy concern, we are not uploading the actual dataset of patient posing expression in image and video format.
In each folder, there are 5 folders for each score. Score 1 represents a high degree of facial paralysis and Score 5 represents a low degree of facial paralysis.
For each sample like "af01_round2_001Camera1" - 
  • a represents the age of 17year old patients, similarly b - 18-29, c - 30-44, d - 45-59, e - 60-69, f - 70

  • f represents male and m represents the female candidate

  • 001 represents the expression type (000-EP0, 001-EP1..... and 009-EP9)

  • Camera1 represent view 0 degree (camera2 30 degree.....camera7-180 deg

 

The zip file is password protected. Kindly email to the author for the password and the consent form.

 

For more detail or query please email to cs14resch11006@iith.ac.in or 786.nazil@gmail.com

Comments

cc

Submitted by yufei wu on Tue, 01/26/2021 - 01:59

where is the dataset ,i am unable to find it.plz help???

Submitted by sahil singla on Thu, 09/07/2023 - 02:56

Dataset Files

    Files have not been uploaded for this dataset

    Documentation

    AttachmentSize
    File About FPD dataset47.32 KB