Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

MULTIGAP: MULTI-POOLED INCEPTION NETWORK WITH TEXT AUGMENTATION FOR AESTHETIC PREDICTION OF PHOTOGRAPHS

Abstract: 

With the advent of deep learning, convolutional neural networks have solved many imaging problems to a large extent. However, it remains to be seen if the image “bottleneck” can be unplugged by harnessing complementary sources of data. In this paper, we present a new approach to image aesthetic evaluation that learns both visual and textual features simultaneously. Our network extracts visual features by appending global average pooling blocks on multiple inception modules (MultiGAP), while textual features from associated user comments are learned from a recurrent neural network. Experimental results show that the proposed method is capable of achieving state-of-the-art performance on the AVA / AVA Comments datasets. We also demonstrate the capability of our approach in visualizing aesthetic activations.

up
0 users have voted:

Paper Details

Authors:
Yong-Lian Hii, John See, Magzhan Kairanbay, Lai-Kuan Wong
Submitted On:
15 September 2017 - 4:38am
Short Link:
Type:
Presentation Slides
Event:
Presenter's Name:
Magzhan Kairanbay
Paper Code:
3306
Document Year:
2017
Cite

Document Files

MULTIGAP- MULTI-POOLED INCEPTION NETWORK WITH TEXT AUGMENTATION FOR AESTHETIC PREDICTION OF PHOTOGRAPHS.pdf

(56)

Subscribe

[1] Yong-Lian Hii, John See, Magzhan Kairanbay, Lai-Kuan Wong, "MULTIGAP: MULTI-POOLED INCEPTION NETWORK WITH TEXT AUGMENTATION FOR AESTHETIC PREDICTION OF PHOTOGRAPHS", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/2109. Accessed: Jan. 18, 2020.
@article{2109-17,
url = {http://sigport.org/2109},
author = {Yong-Lian Hii; John See; Magzhan Kairanbay; Lai-Kuan Wong },
publisher = {IEEE SigPort},
title = {MULTIGAP: MULTI-POOLED INCEPTION NETWORK WITH TEXT AUGMENTATION FOR AESTHETIC PREDICTION OF PHOTOGRAPHS},
year = {2017} }
TY - EJOUR
T1 - MULTIGAP: MULTI-POOLED INCEPTION NETWORK WITH TEXT AUGMENTATION FOR AESTHETIC PREDICTION OF PHOTOGRAPHS
AU - Yong-Lian Hii; John See; Magzhan Kairanbay; Lai-Kuan Wong
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/2109
ER -
Yong-Lian Hii, John See, Magzhan Kairanbay, Lai-Kuan Wong. (2017). MULTIGAP: MULTI-POOLED INCEPTION NETWORK WITH TEXT AUGMENTATION FOR AESTHETIC PREDICTION OF PHOTOGRAPHS. IEEE SigPort. http://sigport.org/2109
Yong-Lian Hii, John See, Magzhan Kairanbay, Lai-Kuan Wong, 2017. MULTIGAP: MULTI-POOLED INCEPTION NETWORK WITH TEXT AUGMENTATION FOR AESTHETIC PREDICTION OF PHOTOGRAPHS. Available at: http://sigport.org/2109.
Yong-Lian Hii, John See, Magzhan Kairanbay, Lai-Kuan Wong. (2017). "MULTIGAP: MULTI-POOLED INCEPTION NETWORK WITH TEXT AUGMENTATION FOR AESTHETIC PREDICTION OF PHOTOGRAPHS." Web.
1. Yong-Lian Hii, John See, Magzhan Kairanbay, Lai-Kuan Wong. MULTIGAP: MULTI-POOLED INCEPTION NETWORK WITH TEXT AUGMENTATION FOR AESTHETIC PREDICTION OF PHOTOGRAPHS [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/2109