Scene Recognition Based on Recurrent Memorized Attention Network
Open Access
- 1 December 2020
- journal article
- research article
- Published by MDPI AG in Electronics
- Vol. 9 (12) , 2038
Abstract
We propose a new end-to-end scene recognition framework, called a Recurrent Memorized Attention Network (RMAN) model, which performs object-based scene classification by recurrently locating and memorizing objects in the image. Based on the proposed framework, we introduce a multi-task mechanism that contiguously attends on the different essential objects in a scene image and recurrently performs memory fusion of the features of object focused by an attention model to improve the scene recognition accuracy. The experimental results show that the RMAN model has achieved better classification performance on the constructed dataset and two public scene datasets, surpassing state-of-the-art image scene recognition approaches.Keywords
This publication has 31 references indexed in Scilit:
- CNN-RNN: A Unified Framework for Multi-label Image Classification2016
- Harvesting Discriminative Meta Objects with Deep CNN Features for Scene Classification2015
- ImageNet Large Scale Visual Recognition ChallengeInternational Journal of Computer Vision, 2015
- Learning Important Spatial Pooling Regions for Scene Classification2014
- CNN Features Off-the-Shelf: An Astounding Baseline for Recognition2014
- mCENTRIST: A Multi-Channel Feature Generation Mechanism for Scene CategorizationIEEE Transactions on Image Processing, 2013
- Blocks That Shout: Distinctive Parts for Scene Classification2013
- A tutorial on ν‐support vector machinesApplied Stochastic Models in Business and Industry, 2005
- Modeling the Shape of the Scene: A Holistic Representation of the Spatial EnvelopeInternational Journal of Computer Vision, 2001
- Gradient-based learning applied to document recognitionProceedings of the IEEE, 1998