Talk · Proceedings - International Conference on Image Processing · September 1, 2017
Multigap: Multi-pooled inception network with text augmentation for aesthetic prediction of photographs
Yong-Lian Hii, John See, Magzhan Kairanbay, Lai-Kuan Wong
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.
More by Magzhan Kairanbay
- End-to-End VITS Outperforms Two-Stage Tacotron 2 Synthesis for Kazakh: A Comparative Study on a New Multi-Domain Speech CorpusAlgorithms · 2026
- Vision-Language Temporal Analysis for Illegal Waste Dumping Detection in Surveillance VideosIEEE/CVF Winter Conference on Applications of Computer Vision Workshops · 2026 · Citations: 2
- Real-Time Multi-Parameter Petroleum Coke Quality Monitoring Using Computer Vision and Integrated Anomaly Detection for Aluminum Smelting OperationsInternational Conference on Computer and Applications · 2025
- Multi-Dimensional Tail Classification in Procurement Spend Analysis: A Data-Driven Approach to Strategic Sourcing OptimizationInternational Conference on Computer and Applications · 2025