Image Captioning Research Articles

Leveraging the Jetson Nano AI Kit for Machine Learning in Quantifying Gender Bias in Image Captioning

In an era where artificial intelligence intertwines with ethical concerns, this paper delves into the practical deployment of the Jetson Nano AI kit to assess gender bias in image captioning. As the AI landscape evolves, the Jetson Nano stands out as an effective tool for environments with limited resources, leveraging its edge computing capabilities to reshape AI research. The paper examines the foundational concepts of gender bias quantification and its real-world application using the Jetson Nano, highlighting its versatility in constrained environments. Throughout the exploration, the seamless integration of the Jetson Nano into machine learning processes is detailed, shedding light on crucial optimizations and adjustments. Challenges encountered are also analyzed, offering insights for researchers undertaking similar projects. Ultimately, the research underscores not just Jetson Nano's role in edge computing but also the imperative of confronting gender bias in AI-generated image descriptions. By melding ethical considerations with edge computing, this paper paves the way for a more balanced and effective AI future.

VTG-GPT: Tuning-Free Zero-Shot Video Temporal Grounding with GPT

Video temporal grounding (VTG) aims to locate specific temporal segments from an untrimmed video based on a linguistic query. Most existing VTG models are trained on extensive annotated video-text pairs, a process that not only introduces human biases from the queries but also incurs significant computational costs. To tackle these challenges, we propose VTG-GPT, a GPT-based method for zero-shot VTG without training or fine-tuning. To reduce prejudice in the original query, we employ Baichuan2 to generate debiased queries. To lessen redundant information in videos, we apply MiniGPT-v2 to transform visual content into more precise captions. Finally, we devise the proposal generator and post-processing to produce accurate segments from debiased queries and image captions. Extensive experiments demonstrate that VTG-GPT significantly outperforms SOTA methods in zero-shot settings and surpasses unsupervised approaches. More notably, it achieves competitive performance comparable to supervised methods. The code is available on GitHub.

Perceptual Image Quality Prediction: Are Contrastive Language–Image Pretraining (CLIP) Visual Features Effective?

In recent studies, the Contrastive Language–Image Pretraining (CLIP) model has showcased remarkable versatility in downstream tasks, ranging from image captioning and question-answering reasoning to image–text similarity rating, etc. In this paper, we investigate the effectiveness of CLIP visual features in predicting perceptual image quality. CLIP is also compared with competitive large multimodal models (LMMs) for this task. In contrast to previous studies, the results show that CLIP and other LMMs do not always provide the best performance. Interestingly, our evaluation experiment reveals that combining visual features from CLIP or other LMMs with some simple distortion features can significantly enhance their performance. In some cases, the improvements are even more than 10%, while the prediction accuracy surpasses 90%.

RVAIC: Refined visual attention for improved image captioning

Visual attention has emerged as a prominent approach for improving the effectiveness of image captioning, as it enables the decoder network to focus selectively on the most salient regions in the image content, thereby facilitating the generation of precise and informative captions. Although visual attention achieves the improvement, the small numerical values of its input have a negative impact on its softmax, decreasing its effectiveness. To address this limitation, we propose a refined visual attention (RVA) framework that internally reweights visual attention by leveraging the language context of previously generated words. We first feed the language context into a fully connected layer to obtain appropriate dimensions for the visual features. Then, we use a sigmoid function to obtain a probability distribution to reweight the softmax’s input by applying the multiplication process. Experiments conducted on the MS COCO dataset demonstrate that RVA outperforms traditional visual attention and other existing image captioning methods, highlighting its effectiveness in enhancing the accuracy and informativeness of image captions.

A focus fusion attention mechanism integrated with image captions for knowledge graph-based visual question answering

A focus fusion attention mechanism integrated with image captions for knowledge graph-based visual question answering

Exploring better image captioning with grid features

Nowadays, Artificial Intelligence Generated Content (AIGC) has shown promising prospects in both computer vision and natural language processing communities. Meanwhile, as an essential aspect of AIGC, image to captions has received much more attention. Recent vision-language research is developing from the bulky region visual representations based on object detectors toward more convenient and flexible grid ones. However, this kind of research typically concentrates on image understanding tasks like image classification, with less attention paid to content generation tasks. In this paper, we explore how to capitalize on the expressive features embedded in the grid visual representations for better image captioning. To this end, we present a Transformer-based image captioning model, dubbed FeiM, with two straightforward yet effective designs. We first design the feature queries that consist of a limited set of learnable vectors, which act as the local signals to capture specific visual information from global grid features. Then, taking augmented global grid features and the local feature queries as inputs, we develop a feature interaction module to query relevant visual concepts from grid features, and to enable interaction between the local signal and overall context. Finally, the refined grid visual representations and the linguistic features pass through a Transformer architecture for multi-modal fusion. With the two novel and simple designs, FeiM can fully leverage meaningful visual knowledge to improve image captioning performance. Extensive experiments are performed on the competitive MSCOCO benchmark to confirm the effectiveness of the proposed approach, and the results show that FeiM yields more eminent results than existing advanced captioning models.

Enhancing Image Captioning and Auto-Tagging Through a FCLN with Faster R-CNN Integration

Enhancing Image Captioning and Auto-Tagging Through a FCLN with Faster R-CNN Integration

Egocentric Image Captioning for Privacy-Preserved Passive Dietary Intake Monitoring.

Camera-based passive dietary intake monitoring is able to continuously capture the eating episodes of a subject, recording rich visual information, such as the type and volume of food being consumed, as well as the eating behaviors of the subject. However, there currently is no method that is able to incorporate these visual clues and provide a comprehensive context of dietary intake from passive recording (e.g., is the subject sharing food with others, what food the subject is eating, and how much food is left in the bowl). On the other hand, privacy is a major concern while egocentric wearable cameras are used for capturing. In this article, we propose a privacy-preserved secure solution (i.e., egocentric image captioning) for dietary assessment with passive monitoring, which unifies food recognition, volume estimation, and scene understanding. By converting images into rich text descriptions, nutritionists can assess individual dietary intake based on the captions instead of the original images, reducing the risk of privacy leakage from images. To this end, an egocentric dietary image captioning dataset has been built, which consists of in-the-wild images captured by head-worn and chest-worn cameras in field studies in Ghana. A novel transformer-based architecture is designed to caption egocentric dietary images. Comprehensive experiments have been conducted to evaluate the effectiveness and to justify the design of the proposed architecture for egocentric dietary image captioning. To the best of our knowledge, this is the first work that applies image captioning for dietary intake assessment in real-life settings.

Exploiting Cross-Modal Prediction and Relation Consistency for Semisupervised Image Captioning.

The task of image captioning aims to generate captions directly from images via the automatically learned cross-modal generator. To build a well-performing generator, existing approaches usually need a large number of described images (i.e., supervised image-sentence pairs), requiring a huge effects on manual labeling. However, in real-world applications, a more general scenario is that we only have limited amount of described images and a large number of undescribed images. Therefore, a resulting challenge is how to effectively combine the undescribed images into the learning of cross-modal generator (i.e., semisupervised image captioning). To solve this problem, we propose a novel image captioning method by exploiting the cross-modal prediction and relation consistency (CPRC), which aims to utilize the raw image input to constrain the generated sentence in the semantic space. In detail, considering that the heterogeneous gap between modalities always leads to the supervision difficulty while using the global embedding directly, CPRC turns to transform both the raw image and corresponding generated sentence into the shared semantic space, and measure the generated sentence from two aspects: 1) prediction consistency: CPRC utilizes the prediction of raw image as soft label to distill useful supervision for the generated sentence, rather than employing the traditional pseudo labeling and 2) relation consistency: CPRC develops a novel relation consistency between augmented images and corresponding generated sentences to retain the important relational knowledge. In result, CPRC supervises the generated sentence from both the informativeness and representativeness perspectives, and can reasonably use the undescribed images to learn a more effective generator under the semisupervised scenario. The experiments show that our method outperforms state-of-the-art comparison methods on the MS-COCO "Karpathy" offline test split under complex nonparallel scenarios, for example, CPRC achieves at least 6% improvements on the CIDEr-D score.

ArtCap: A Dataset for Image Captioning of Fine Art Paintings

The image captioning of fine art paintings aims at generating content descriptions for the paintings. Due to the complexity of modeling both image and language, this task usually needs sufficient training data. However, different from photographic image captioning, there are few satisfactory datasets for painting captioning. In this article, we introduce a painting captioning dataset (named the ArtCap dataset), which contains 3606 paintings and five descriptions for each painting. We present the carefully designed construction pipeline of our dataset and further evaluate our dataset from two aspects of annotation quality and application effectiveness, respectively. For the annotation quality, we compare the global characteristics, annotation content, and annotation consistency of our dataset with other painting descriptions datasets. For application effectiveness, we employ our dataset and other painting descriptions datasets to train image captioning models and analyze the captioning performances. The results demonstrate the promising annotation quality and application effectiveness of our dataset.

Adaptive Semantic-Enhanced Transformer for Image Captioning.

In the research on image captioning, rich semantic information is very important for generating critical caption words as guiding information. However, semantic information from offline object detectors involves many semantic objects that do not appear in the caption, thereby bringing noise into the decoding process. To produce more accurate semantic guiding information and further optimize the decoding process, we propose an end-to-end adaptive semantic-enhanced transformer (AS-Transformer) model for image captioning. For semantic enhancement information extraction, we propose a constrained weaklysupervised learning (CWSL) module, which reconstructs the semantic object's probability distribution detected by the multiple instances learning (MIL) through a joint loss function. These strengthened semantic objects from the reconstructed probability distribution can better depict the semantic meaning of images. Also, for semantic enhancement decoding, we propose an adaptive gated mechanism (AGM) module to adjust the attention between visual and semantic information adaptively for the more accurate generation of caption words. Through the joint control of the CWSL module and AGM module, our proposed model constructs a complete adaptive enhancement mechanism from encoding to decoding and obtains visual context that is more suitable for captions. Experiments on the public Microsoft Common Objects in COntext (MSCOCO) and Flickr30K datasets illustrate that our proposed AS-Transformer can adaptively obtain effective semantic information and adjust the attention weights between semantic and visual information automatically, which achieves more accurate captions compared with semantic enhancement methods and outperforms state-of-the-art methods.

Multimodal Fake News Analysis Based on Image–Text Similarity

With the fast and extensive development of computer vision techniques, multimodal analyses are utilized more frequently for online fake news detection. To better understand the image–text relationship and its role in fake news detection, in this article, we proposed and evaluated four image–text similarities, namely, textual similarity, semantic similarity, contextual similarity, and post-training similarity. The textual and semantic similarities indicate the original image–text similarities in terms of the text information and image caption information. The contextual similarity reflects the image–text similarity in the format of meaningful named entities. The post-training similarity demonstrates how image–text similarity involves before and after a fake news detection model is trained. By evaluating the proposed similarity measurements on three real-world datasets, we find that fake news image–text similarity is higher than real news image–text similarity in most of the cases. Furthermore, the comparison of models’ performance further validates the significance of visual information in online fake news detection. These findings may be considered as the fundamental logic to explain the original purpose of fake news creation and can be used as influential features for improving models’ performance in the future.

IMAGE CAPTIONING WITH AUDIO INTEGRATION

Investigating the synergy between image captioning and audio integration, our study employs an advanced Encoder-Decoder framework. Integrating visual and auditory features, the model enhances caption precision. Through extensive experimentation, our results reveal a notable improvement in contextual understanding compared to conventional image captioning models. This research underscores the potential of multimodal approaches for enriching multimedia applications, particularly in contexts where comprehensive content comprehension and accessibility are paramount. Key Words: Deep Learning, Natural Language Processing, Multi Modal Attention Mechanism, Image Captioning.

DIC-Transformer: interpretation of plant disease classification results using image caption generation technology

Disease image classification systems play a crucial role in identifying disease categories in the field of agricultural diseases. However, current plant disease image classification methods can only predict the disease category and do not offer explanations for the characteristics of the predicted disease images. Due to the current situation, this paper employed image description generation technology to produce distinct descriptions for different plant disease categories. A two-stage model called DIC-Transformer, which encompasses three tasks (detection, interpretation, and classification), was proposed. In the first stage, Faster R-CNN was utilized to detect the diseased area and generate the feature vector of the diseased image, with the Swin Transformer as the backbone. In the second stage, the model utilized the Transformer to generate image captions. It then generated the image feature vector, which is weighted by text features, to improve the performance of image classification in the subsequent classification decoder. Additionally, a dataset containing text and visualizations for agricultural diseases (ADCG-18) was compiled. The dataset contains images of 18 diseases and descriptive information about their characteristics. Then, using the ADCG-18, the DIC-Transformer was compared to 11 existing classical caption generation methods and 10 image classification models. The evaluation indicators for captions include Bleu1–4, CiderD, and Rouge. The values of BLEU-1, CIDEr-D, and ROUGE were 0.756, 450.51, and 0.721. The results of DIC-Transformer were 0.01, 29.55, and 0.014 higher than those of the highest-performing comparison model, Fc. The classification evaluation metrics include accuracy, recall, and F1 score, with accuracy at 0.854, recall at 0.854, and F1 score at 0.853. The results of DIC-Transformer were 0.024, 0.078, and 0.075 higher than those of the highest-performing comparison model, MobileNetV2. The results indicate that the DIC-Transformer outperforms other comparison models in classification and caption generation.

Machine-to-Machine Visual Dialoguing with ChatGPT for Enriched Textual Image Description

Image captioning is a technique that enables the automatic extraction of natural language descriptions about the contents of an image. On the one hand, information in the form of natural language can enhance accessibility by reducing the expertise required to process, analyze, and exploit remote sensing images, while on the other, it provides a direct and general form of communication. However, image captioning is usually restricted to a single sentence, which barely describes the rich semantic information that typically characterizes remote sensing (RS) images. In this paper, we aim to move one step forward by proposing a captioning system that, mimicking human behavior, adopts dialogue as a tool to explore and dig for information, leading to more detailed and comprehensive descriptions of RS scenes. The system relies on a questions–answers scheme fed by a query image and summarizes the dialogue content with ChatGPT. Experiments carried out on two benchmark remote sensing datasets confirm the potential of such an approach in the context of semantic information mining. Strengths and weaknesses are highlighted and discussed, as well as some possible future developments.

Exploring Visual Relationships via Transformer-based Graphs for Enhanced Image Captioning

Image captioning (IC), bringing vision to language, has drawn extensive attention. A crucial aspect of IC is the accurate depiction of visual relations among image objects. Visual relations encompass two primary facets: content relations and structural relations. Content relations, which comprise geometric positions content (i.e., distances and sizes) and semantic interactions content (i.e., actions and possessives), unveil the mutual correlations between objects. In contrast, structural relations pertain to the topological connectivity of object regions. Existing Transformer-based methods typically resort to geometric positions to enhance the visual relations, yet only using the shallow geometric content is unable to precisely cover actional content correlations and structural connection relations. In this article, we adopt a comprehensive perspective to examine the correlations between objects, incorporating both content relations (i.e., geometric and semantic relations) and structural relations, with the aim of generating plausible captions. To achieve this, first, we construct a geometric graph from bounding box features and a semantic graph from the scene graph parser to model the content relations. Innovatively, we construct a topology graph that amalgamates the sparsity characteristics of the geometric and semantic graphs, enabling the representation of image structural relations. Second, we propose a novel unified approach to enrich image relation representations by integrating semantic, geometric, and structural relations into self-attention. Finally, in the language decoding stage, we further leverage the semantic relation as prior knowledge to generate accurate words. Extensive experiments on MS-COCO dataset demonstrate the effectiveness of our model, with improvements of CIDEr from 128.6% to 136.6%. Codes have been released at https://github.com/CrossmodalGroup/ER-SAN/tree/main/VG-Cap .

An efficient automated image caption generation by the encoder decoder model

An efficient automated image caption generation by the encoder decoder model

ReverseGAN: An intelligent reverse generative adversarial networks system for complex image captioning generation

Towards the inclusion of complex semantic relational images, we propose an intelligent Reverse Generative Adversarial Network (ReverseGAN) with generative task guidance to build an image caption system. The system utilizes regenerated images to learn the concept of image caption generation, using a generative adversarial network as the overall framework of the model. The generative network uses a graph convolutional neural network to encode the images and constructs a decoder model that converts image vectors into captions. The reverse text-to-image task serves as a discriminator model. The discriminator uses a text embedding module to map the text descriptions generated by the generator to local word-level features and global sentence features. In addition, the introduced cascading attention module utilizes the generated embedding vectors to generate images on a coarse-to-fine scale, incorporating both global and local features of the text to ensure that the generated images are closer to the original images. Our model outperforms current state-of-the-art methods in BLEU, METEOR, and ROUGE metrics when experimented on the MSCOCO dataset.

Images, Words, and Imagination: Accessible Descriptions to Support Blind and Low Vision Art Exploration and Engagement.

The lack of accessible information conveyed by descriptions of art images presents significant barriers for people with blindness and low vision (BLV) to engage with visual artwork. Most museums are not able to easily provide accessible image descriptions for BLV visitors to build a mental representation of artwork due to vastness of collections, limitations of curator training, and current measures for what constitutes effective automated captions. This paper reports on the results of two studies investigating the types of information that should be included to provide high-quality accessible artwork descriptions based on input from BLV description evaluators. We report on: (1) a qualitative study asking BLV participants for their preferences for layered description characteristics; and (2) an evaluation of several current models for image captioning as applied to an artwork image dataset. We then provide recommendations for researchers working on accessible image captioning and museum engagement applications through a focus on spatial information access strategies.

Overcoming the Challenge of Cyberbullying Detection in Images: A Deep Learning Approach with Image Captioning and OCR Integration

Overcoming the Challenge of Cyberbullying Detection in Images: A Deep Learning Approach with Image Captioning and OCR Integration