What is it about?

The paper presents a review of three deep learning algorithms for Automatic Image Annotation (AIA), focusing on a proposed Convolutional Neural Network (CNN) model. The model aims to generate semantic concepts from visual features of images, automatically assigning labels to new images for efficient retrieval. The paper discusses challenges in image annotation, the use of CNN for AIA, and the future scope of the proposed model.

Featured Image

Why is it important?

This research is important because it proposes an automatic image annotation (AIA) model that assigns semantic labels to new images, facilitating efficient image retrieval and organization. The paper presents a deep learning approach based on Convolutional Neural Networks (CNN) to address challenges in assigning labels to images and speed up the image retrieval process. Key Takeaways: 1. The paper proposes an AIA model that assigns semantic labels to new images for efficient image retrieval. 2. The model is based on deep learning, specifically Convolutional Neural Networks (CNN), and uses a dataset of Corel-5K, ESP Game, IAPR TC-12, NUSWIDE, and MS-COCO. 3. The method utilizes a CNN-RNN model, with Back Propagation Through Time (BPTT) for adjusting the neural network weights, and supervised learning for correcting errors. 4. The proposed model is implemented using Python 3.7, with potential for enhancing results through modifications. 5. The research aims to bridge the semantic gap between low-level visual image information and high-level semantic notions perceived by people. 6. Future work will focus on addressing annotation and sorting issues using current approaches for efficient image retrieval.

AI notice

Some of the content on this page has been created using generative AI.

Read the Original

This page is a summary of: Automatic image annotation system to images retrieval based on deep learning technique, January 2023, American Institute of Physics,
DOI: 10.1063/5.0167678.
You can read the full text:

Read

Contributors

The following have contributed to this page