TY  - JOUR
Y1  - 2017///
A1  - Pampouchidou, A.
A1  - Pediaditis, M.
A1  - Maridaki, A.
A1  - Awais, M.
A1  - Vazakopoulou, C.-M.
A1  - Sfakianakis, S.
A1  - Tsiknakis, M.
A1  - Simos, P.
A1  - Marias, K.
A1  - Yang, F.
A1  - Meriaudeau, F.
ID  - scholars8102
TI  - Quantitative comparison of motion history image variants for video-based depression assessment
SN  - 16875176
IS  - 1
KW  - Classification (of information); Image processing; Learning systems; Neural networks
KW  -  Affective Computing; Depression assessment; Facial images; Facial landmark; Motion history images
KW  -  Video streaming
N1  - cited By 13
JF  - Eurasip Journal on Image and Video Processing
VL  - 2017
N2  - Depression is the most prevalent mood disorder and a leading cause of disability worldwide. Automated video-based analyses may afford objective measures to support clinical judgments. In the present paper, categorical depression assessment is addressed by proposing a novel variant of the Motion History Image (MHI) which considers Gabor-inhibited filtered data instead of the original image. Classification results obtained with this method on the AVECâ??14 dataset are compared to those derived using (a) an earlier MHI variant, the Landmark Motion History Image (LMHI), and (b) the original MHI. The different motion representations were tested in several combinations of appearance-based descriptors, as well as with the use of convolutional neural networks. The F1 score of 87.4 achieved in the proposed work outperformed previously reported approaches. © 2017, The Author(s).
PB  - Springer International Publishing
AV  - none
UR  - https://www.scopus.com/inward/record.uri?eid=2-s2.0-85029085354&doi=10.1186%2fs13640-017-0212-3&partnerID=40&md5=dcb9ebb1cff424e3d8e6d1668ae60349
ER  -