Content-based image retrieval with compact deep convolutional features

Abstract

Convolutional neural networks (CNNs) with deep learning have recently achieved a remarkable success with a superior performance in computer vision applications. Most of CNN-based methods extract image features at the last layer using a single CNN architecture with orderless quantization approaches, which limits the utilization of intermediate convolutional layers for identifying image local patterns. As one of the first works in the context of content-based image retrieval (CBIR), this paper proposes a new bilinear CNN-based architecture using two parallel CNNs as feature extractors. The activations of convolutional layers are directly used to extract the image features at various image locations and scales. The network architecture is initialized by deep CNNs sufficiently pre-trained on large generic image dataset then fine-tuned for the CBIR task. Additionally, an efficient bilinear root pooling is proposed and applied to the low-dimensional pooling layer to reduce the dimension of image features to compact but high discriminative image descriptors. Finally, an endto-end training with backpropagation is performed to fine-tune the final architecture and to learn its parameters for the image retrieval task. The experimental results achieved on three standard benchmarking image datasets demonstrate the outstanding performance of the proposed architecture at extracting and learning complex features for the CBIR task without prior knowledge about the semantic meta-data of images. For instance, using a very compact image vector of 16-length, we achieve retrieval accuracy 95.7% (mAP) on Oxford5K and 88.6% on Oxford105K; which outperforms the best results reported by state-of-the-art approaches. Additionally, a noticeable reduction is attained in the required extraction time for image features and the memory size required for storage.

Deep Learning Using Symmetry, FAST Scores, Shape-Based Filtering and Spatial Mapping Integrated with CNN for Large Scale Image Retrieval

“…These machines and networks are used for target detection and object recognition [8]. Deep learning is implemented using learning features with CNN architecture in [47].…”

Content based Image Retrieval System using Combination of Color and Shape Features, and Siamese Neural Network

“…The experiments have shown effective CBIR system with the combination of features than using individual features. Many recent works [8,9,10,11,12] have used the convolutional neural network for extraction of features from the images and store them. These methods have shown superior performances than the earlier methods of feature extraction.…”

An Novel Framework For Content Based Image Retrieval With Quality Assessment System using Optimal Deep Convolution Neural Network

“…The features of CNN and the bag of words aggregation model is presented in [13] whereas the complementary benefits of the CNN features of diverse layers are introduced in [14] and outperforms the integration of many layers. In [15], a bilinear CNN-based architecture is developed in CBIR field where a bilinear root pooling is presented for projecting the features filtered from two parallel CNN models to a low dimension and the resultant model undergo training on the image retrieval dataset by the use of unsupervised training.…”