Content and Context Features for Scene Image Representation

  title={Content and Context Features for Scene Image Representation},
  author={Chiranjibi Sitaula and Sunil Aryal and Yong Xiang and Anish Basnet and Xuequan Lu},
  journal={Knowl. Based Syst.},
Recent Advances in Scene Image Representation and Classification
This survey provides in-depth insights and applications of recent scene image representation methods for traditional Computer Vision (CV)-based methods, Deep Learning (DL) based methods, and Search Engine (SE)-based methods.
Traditional Ceramic Sculpture Feature Recognition Based on the Machine Learning Algorithm
  • XiaoLei Zhu
  • Computer Science
    Wireless Communications and Mobile Computing
  • 2022
The recognition algorithm designed in this paper has high accuracy in identifying the traditional ceramic sculpture features, can effectively suppress the line noise, and has a short recognition time overhead, which has a certain feasibility.
Scene Classification using Regional and Nearest Neighbors of Local features
The results show that the proposed models produce better results than the existing bag of words model and context models from literature in terms of classification accuracy.
Fusion of multi-scale bag of deep visual words features of chest X-ray images to detect COVID-19 infection
A new multi-scale bag of deep visual words (MBoDVW) features, which exploits three different scales of the 4th pooling layer’s output feature map achieved from VGG-16 model is proposed.
A Multiclass Nonparallel Parametric-Margin Support Vector Machine
This paper extends TPMSVM for multiclass classification and proposes a novel K multiclass nonparallel parametric-margin support vector machine (MNP-KSVC), which utilizes a hybrid classification and regression loss joined with the regularization to formulate its optimization model.
Vector representation based on a supervised codebook for Nepali documents classification
This article proposes a novel document representation method based on a supervised codebook to represent the Nepali documents, where the codebook contains only semantic tokens without outliers and yields the best classification accuracy on three datasets and a comparable accuracy on the fourth dataset.
New bag of deep visual words based features to classify chest x-ray images for COVID-19 diagnosis
A new BoVW method over deep features, called Bag of Deep Visual Words (BoDVW), by removing the feature map normalization step and adding the deep featuresnormalization step on the raw feature maps is proposed to preserve the semantics of each feature map that may have important clues to differentiate COVID-19 from Pneumonia.


When Naïve Bayes Nearest Neighbors Meet Convolutional Neural Networks
This paper proposes a framework that addresses all issues of Naive Bayes Nearest Neighbor (NBNN)-based classifiers, thus bringing back NBNNs on the map, and addresses simultaneously the second and third by proposing a scalable version of Naïve Bayes Non-linear Learning (NBNL).
HDF: Hybrid Deep Features for Scene Image Representation
This paper proposes a novel type of features – hybrid deep features, for scene images which exploit both object-based and scene-based features at two levels: part image level and whole image level, which produces a total number of four types of deep features.
Coordinate CNNs and LSTMs to categorize scene images with multi-views and multi-levels of abstraction
Places: A 10 Million Image Database for Scene Recognition
The Places Database is described, a repository of 10 million scene photographs, labeled with scene semantic categories, comprising a large and diverse list of the types of environments encountered in the world, using the state-of-the-art Convolutional Neural Networks as baselines, that significantly outperform the previous approaches.
Bag of Surrogate Parts: one inherent feature of deep CNNs
This paper develops a new feature from convolutional layers, called Bag of Surrogate Parts (BoSP), and its spatial variant, Spatial BoSP, which is efficient, has no tuning parameters, and could generate low-dimensional, highly discriminative features.
Sift Descriptors Modeling and Application in Texture Image Classification
A new statistical model for describing real textured images based on the observation that the Scale-Invariant Feature Transform (SIFT) descriptors extracted from a given image can be properly modeled by the Gamma distribution is presented.
Places: An Image Database for Deep Scene Understanding
The Places Database is described, a repository of 10 million scene photographs, labeled with scene semantic categories and attributes, comprising a quasi-exhaustive list of the types of environments encountered in the world.
Learning Important Spatial Pooling Regions for Scene Classification
This work addresses the false response influence problem when learning and applying discriminative parts to construct the mid-level representation in scene classification by learning important spatial pooling regions along with their appearance and achieves state-of-the-art performance on several datasets.
Very Deep Convolutional Networks for Large-Scale Image Recognition
This work investigates the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting using an architecture with very small convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers.