Image-text Retrieval Research Articles

Retraction Note: Simulation of cross-modal image-text retrieval algorithm under convolutional neural network structure and hash method

Retraction Note: Simulation of cross-modal image-text retrieval algorithm under convolutional neural network structure and hash method

Relation-Aggregated Cross-Graph Correlation Learning for Fine-Grained Image-Text Retrieval.

Fine-grained image-text retrieval has been a hot research topic to bridge the vision and languages, and its main challenge is how to learn the semantic correspondence across different modalities. The existing methods mainly focus on learning the global semantic correspondence or intramodal relation correspondence in separate data representations, but which rarely consider the intermodal relation that interactively provide complementary hints for fine-grained semantic correlation learning. To address this issue, we propose a relation-aggregated cross-graph (RACG) model to explicitly learn the fine-grained semantic correspondence by aggregating both intramodal and intermodal relations, which can be well utilized to guide the feature correspondence learning process. More specifically, we first build semantic-embedded graph to explore both fine-grained objects and their relations of different media types, which aim not only to characterize the object appearance in each modality, but also to capture the intrinsic relation information to differentiate intramodal discrepancies. Then, a cross-graph relation encoder is newly designed to explore the intermodal relation across different modalities, which can mutually boost the cross-modal correlations to learn more precise intermodal dependencies. Besides, the feature reconstruction module and multihead similarity alignment are efficiently leveraged to optimize the node-level semantic correspondence, whereby the relation-aggregated cross-modal embeddings between image and text are discriminatively obtained to benefit various image-text retrieval tasks with high retrieval performance. Extensive experiments evaluated on benchmark datasets quantitatively and qualitatively verify the advantages of the proposed framework for fine-grained image-text retrieval and show its competitive performance with the state of the arts.

Integrating listwise ranking into pairwise-based image-text retrieval

Image-Text Retrieval (ITR) is essentially a ranking problem. Given a query caption, the goal is to rank candidate images by relevance, from large to small. The current ITR datasets are constructed in a pairwise manner. Image-text pairs are annotated as positive or negative. Correspondingly, ITR models mainly use pairwise losses, such as triplet loss, to learn to rank. Pairwise-based ITR increases positive pair similarity while decreasing negative pair similarity indiscriminately. However, the relevance between dissimilar negative pairs is different. Pairwise annotations cannot reflect this difference in relevance. In the current datasets, pairwise annotations miss many correlations. There are many potential positive pairs among the pairs labeled as negative. Pairwise-based ITR can only rank positive samples before negative samples, but cannot rank negative samples by relevance. In this paper, we integrate listwise ranking into conventional pairwise-based ITR. Listwise ranking optimizes the entire ranking list based on relevance scores. Specifically, we first propose a Relevance Score Calculation (RSC) module to calculate the relevance score of the entire ranked list. Then we choose the ranking metric, Normalised Discounted Cumulative Gain (NDCG), as the optimization objective. We apply a metric smoothing method named Smooth-NDCG (S-NDCG) to ITR, which transforms the non-differentiable NDCG into a differentiable listwise loss. Our listwise ranking approach can be plug-and-play integrated into current pairwise-based ITR models. Experiments on ITR benchmarks show that integrating listwise ranking can improve the performance of current ITR models and provide more user-friendly retrieval results. The code is available at https://github.com/AAA-Zheng/Listwise_ITR.

Continual learning for cross-modal image-text retrieval based on domain-selective attention

Cross-modal image-text retrieval (CMITR) has been a high-value research topic for more than a decade. In most of the previous studies, the data for all tasks are trained as a single set. However, in reality, a more likely scenario is that the dataset has multiple tasks and trains them in sequence. The consequence is the limited ability to memorize the old task once a new task arrives; in other words, catastrophic forgetting. To solve this issue, this paper proposes a novel continual learning for cross-modal image-text retrieval (CLCMR) method to alleviate catastrophic forgetting. We construct a multilayer domain-selective attention (MDSA) based network to obtain knowledge from task-relevant and domain-specific attention levels. Moreover, a memory factor has been designed to achieve weight regularization, and a novel memory loss function is utilized to constrain MDSA. The extensive experimental results from multiple datasets (Wikipedia, Pascal Sentence, and PKU XMedianet datasets) demonstrate that CLCMR can effectively alleviate catastrophic forgetting and achieve a superior continual learning ability compared with the state-of-the-art methods.

USER: Unified Semantic Enhancement With Momentum Contrast for Image-Text Retrieval.

As a fundamental and challenging task in bridging language and vision domains, Image-Text Retrieval (ITR) aims at searching for the target instances that are semantically relevant to the given query from the other modality, and its key challenge is to measure the semantic similarity across different modalities. Although significant progress has been achieved, existing approaches typically suffer from two major limitations: (1) It hurts the accuracy of the representation by directly exploiting the bottom-up attention based region-level features where each region is equally treated. (2) It limits the scale of negative sample pairs by employing the mini-batch based end-to-end training mechanism. To address these limitations, we propose a Unified Semantic Enhancement Momentum Contrastive Learning (USER) method for ITR. Specifically, we delicately design two simple but effective Global representation based Semantic Enhancement (GSE) modules. One learns the global representation via the self-attention algorithm, noted as Self-Guided Enhancement (SGE) module. The other module benefits from the pre-trained CLIP module, which provides a novel scheme to exploit and transfer the knowledge from an off-the-shelf model, noted as CLIP-Guided Enhancement (CGE) module. Moreover, we incorporate the training mechanism of MoCo into ITR, in which two dynamic queues are employed to enrich and enlarge the scale of negative sample pairs. Meanwhile, a Unified Training Objective (UTO) is developed to learn from mini-batch based and dynamic queue based samples. Extensive experiments on the benchmark MSCOCO and Flickr30K datasets demonstrate the superiority of both retrieval accuracy and inference efficiency. For instance, compared with the existing best method NAAF, the metric R@1 of our USER on the MSCOCO 5K Testing set is improved by 5% and 2.4% on caption retrieval and image retrieval without any external knowledge or pre-trained model while enjoying over 60 times faster inference speed. Our source code will be released at https://github.com/zhangy0822/USER.

GSSF: Generalized Structural Sparse Function for Deep Cross-Modal Metric Learning.

Cross-modal metric learning is a prominent research topic that bridges the semantic heterogeneity between vision and language. Existing methods frequently utilize simple cosine or complex distance metrics to transform the pairwise features into a similarity score, which suffers from an inadequate or inefficient capability for distance measurements. Consequently, we propose a Generalized Structural Sparse Function to dynamically capture thorough and powerful relationships across modalities for pair-wise similarity learning while remaining concise but efficient. Specifically, the distance metric delicately encapsulates two formats of diagonal and block-diagonal terms, automatically distinguishing and highlighting the cross-channel relevancy and dependency inside a structured and organized topology. Hence, it thereby empowers itself to adapt to the optimal matching patterns between the paired features and reaches a sweet spot between model complexity and capability. Extensive experiments on cross-modal and two extra uni-modal retrieval tasks (image-text retrieval, person re-identification, fine-grained image retrieval) have validated its superiority and flexibility over various popular retrieval frameworks. More importantly, we further discover that it can be seamlessly incorporated into multiple application scenarios, and demonstrates promising prospects from Attention Mechanism to Knowledge Distillation in a plug-and-play manner.

Towards Fast and Accurate Image-Text Retrieval With Self-Supervised Fine-Grained Alignment

Image-text retrieval requires the system to bridge the heterogenous gap between vision and language for accurate retrieval while keeping the network lightweight-enough for efficient retrieval. Existing trade-off solutions mainly study from the view of incorporating cross-modal interactions with the independent-embedding framework or leveraging stronger pre-trained encoders, which still demand time-consuming similarity measurement or heavyweight model structure in the retrieval stage. In this work, we propose an image-text alignment module SelfAlign on top of the independent-embedding framework, which improves the retrieval accuracy while maintains the retrieval efficiency without extra supervision. SelfAlign contains two collaborative sub-modules that force image-text alignment at both the concept level and context level by self-supervised contrastive learning. It doesn't require cross-modal embedding interactions during training while maintaining independent image and text encoders during retrieval. With comparable time cost, SelfAlign consistently boosts the accuracy of state-of-the-art non-pre-training independent-embedding models respectively by 9.1%, 4.2%, and 6.6% in terms of R@sum score on Flickr30K, MS-COCO 1K and MS-COCO 5K datasets. The retrieval accuracy also outperforms most existing interactive-embedding models with orders of magnitude decrease in retrieval time. The source code is available at: https://github.com/Zjamie813/SelfAlign.

Prompt-Based Learning for Unpaired Image Captioning

Unpaired Image Captioning (UIC) has been developed to learn image descriptions from unaligned vision-language sample pairs. Existing works usually tackle this task using adversarial learning and visual concept reward based on reinforcement learning. However, these existing works were only able to learn limited cross-domain information in vision and language domains, which restrains the captioning performance of UIC. Inspired by the success of Vision-Language Pre-Trained Models (VL-PTMs) in this research, we attempt to infer the cross-domain cue information about a given image from the large VL-PTMs for the UIC task. This research is also motivated by recent successes of prompt learning in many downstream multi-modal tasks, including image-text retrieval and vision question answering. In this work, a semantic prompt is introduced and aggregated with visual features for more accurate caption prediction under the adversarial learning framework. In addition, a metric prompt is designed to select high-quality pseudo image-caption samples obtained from the basic captioning model and refine the model in an iterative manner. Extensive experiments on the COCO and Flickr30 K datasets validate the promising captioning ability of the proposed model. We expect that the proposed prompt-based UIC model will stimulate a new line of research for the VL-PTMs based captioning.

Towards Efficient and Accurate Remote Sensing Image-Text Retrieval with a Coarse-to-Fine Approach

Towards Efficient and Accurate Remote Sensing Image-Text Retrieval with a Coarse-to-Fine Approach

Spatial–Channel Attention Transformer With Pseudo Regions for Remote Sensing Image-Text Retrieval

Spatial–Channel Attention Transformer With Pseudo Regions for Remote Sensing Image-Text Retrieval

Feature First: Advancing Image-Text Retrieval Through Improved Visual Features

Feature First: Advancing Image-Text Retrieval Through Improved Visual Features

A Precise Framework for Rice Leaf Disease Image-Text Retrieval Using FHTW-Net.

Cross-modal retrieval for rice leaf diseases is crucial for prevention, providing agricultural experts with data-driven decision support to address disease threats and safeguard rice production. To overcome the limitations of current crop leaf disease retrieval frameworks, we focused on four common rice leaf diseases and established the first cross-modal rice leaf disease retrieval dataset (CRLDRD). We introduced cross-modal retrieval to the domain of rice leaf disease retrieval and introduced FHTW-Net, a framework for rice leaf disease image-text retrieval. To address the challenge of matching diverse image categories with complex text descriptions during the retrieval process, we initially employed ViT and BERT to extract fine-grained image and text feature sequences enriched with contextual information. Subsequently, two-way mixed self-attention (TMS) was introduced to enhance both image and text feature sequences, with the aim of uncovering important semantic information in both modalities. Then, we developed false-negative elimination-hard negative mining (FNE-HNM) strategy to facilitate in-depth exploration of semantic connections between different modalities. This strategy aids in selecting challenging negative samples for elimination to constrain the model within the triplet loss function. Finally, we introduced warm-up bat algorithm (WBA) for learning rate optimization, which improves the model's convergence speed and accuracy. Experimental results demonstrated that FHTW-Net outperforms state-of-the-art models. In image-to-text retrieval, it achieved R@1, R@5, and R@10 accuracies of 83.5%, 92%, and 94%, respectively, while in text-to-image retrieval, it achieved accuracies of 82.5%, 98%, and 98.5%, respectively. FHTW-Net offers advanced technical support and algorithmic guidance for cross-modal retrieval of rice leaf diseases.

Cross-Modal Remote Sensing Image-Text Retrieval via Context and Uncertainty-Aware Prompt.

The cross-modal remote sensing image-text retrieval (CMRSITR) is a lively research topic in the remote sensing (RS) community. Benefiting from the large pretrained image-text models, many successful CMRSITR methods have been proposed in recent years. Although their performance is attractive, there are still some challenges. First, fine-tuning large pretrained models requires a significant amount of computational resources. Second, most large models are pretrained by natural images, which reduces their effectiveness in processing RS images. To tackle these challenges, we propose a new CMRSITR network named context and uncertainty-aware prompt (CUP). First, prompt tuning theory is introduced into CUP to eliminate the burden of optimization resources. By training the prompt tokens rather than all parameters, the large model's knowledge can be transferred to CMRSITR tasks with small trainable parameters. Second, considering the differences between natural-image-based prior clues and RS images, apart from adopting the free-prompt tokens, we develop a prompt generation module (PGM) to produce the RS-oriented prompt tokens. The specific prompt tokens are rich in object-level messages of RS images, which help CUP narrow the gaps between natural large models and RS images. Third, we further design an uncertainty estimation module (UEM) to whittle down the uncertainties caused by the model and data. This way, can not only the semantic misalignment and intraclass diversity imbalance problems be mitigated but also the RS clues can be deeply explored. Competitive experimental results counted on three public benchmark datasets demonstrate that our CUP can achieve competitive performance in the CMRSITR task compared with many existing methods. Our source codes are available at: https://github.com/TangXu-Group/Cross-modal-remote-sensing-image-and-text-retrieval-models/tree/main/CUP.

Cross-modal fabric image-text retrieval based on convolutional neural network and TinyBERT

Cross-modal fabric image-text retrieval based on convolutional neural network and TinyBERT

Flexible graph-based attention and pooling network for image-text retrieval

Flexible graph-based attention and pooling network for image-text retrieval

Revisiting Hard Negative Mining in Contrastive Learning for Visual Understanding

Efficiently mining and distinguishing hard negatives is the key to Contrastive Learning (CL) in various visual understanding tasks. By properly emphasizing the penalty of hard negatives, Hard Negative Mining (HNM) can improve the CL performance. However, there is no method to quantitatively analyze the penalty strength of hard negatives, which makes training difficult to converge. In this paper, we propose a method for measuring and controlling the penalty strength. We first define a penalty strength metric to provides a quantitative analysis tool for HNM. Then, we propose a Triplet loss with Penalty Strength Control (T-PSC), which can balance the penalty strength of hard negatives and the difficulty of model optimization. In order to verify the effectiveness of the proposed T-PSC method in different modalities, we applied it to two visual understanding tasks: Image–Text Retrieval (ITR) for multi-model processing, and Temporal Action Localization (TAL) for video processing. T-PSC can be applied to existing ITR and TAL models in a plug-and-play manner without any changes. Experiments combined with existing models show that a reasonable control of the penalty strength can speed up training and improve the performance on higher-level tasks.

Adequate alignment and interaction for cross-modal retrieval

Cross-modal retrieval has attracted widespread attention in many cross-media similarity search applications, especially image-text retrieval in the fields of computer vision and natural language processing. Recently, visual and semantic embedding (VSE) learning has shown promising improvements on image-text retrieval tasks. Most existing VSE models employ two unrelated encoders to extract features, then use complex methods to contextualize and aggregate those features into holistic embeddings. Despite recent advances, existing approaches still suffer from two limitations: 1) without considering intermediate interaction and adequate alignment between different modalities, these models cannot guarantee the discriminative ability of representations; 2) existing feature aggregators are susceptible to certain noisy regions, which may lead to unreasonable pooling coefficients and affect the quality of the final aggregated features. To address these challenges, we propose a novel cross-modal retrieval model containing a well-designed alignment module and a novel multimodal fusion encoder, which aims to learn adequate alignment and interaction on aggregated features for effectively bridging the modality gap. Experiments on Microsoft COCO and Flickr30k datasets demonstrates the superiority of our model over the state-of-the-art methods.

Augmented reality, deep learning and vision-language query system for construction worker safety

Low situational awareness contributes to safety incidents in construction. Existing Deep Learning (DL)-based applications lack the capability to provide context-specific and interactive feedback that is essential for workers to fully understand their surrounding environments. This paper proposes the Visual Construction Safety Query (VCSQ) system. The system encompasses real-time Image Captioning (IC), safety-centric Visual Question Answering (VQA), and keyword-based Image-Text Retrieval (ITR), integrated with head-mounted Augmented Reality (AR) devices. System validation includes benchmarks and real-world images. The ITR module posted high recall rates of 0.801 and 0.835 for Recall@5 and @10. The VQA module achieved an 89.7% accuracy rate, and the IC module had a SPICE score of 0.449. Feasibility tests and surveys confirmed the system's practical advantages in different construction scenarios. This study establishes an integration roadmap adaptable to future advancements in interactive DL and immersive AR.

Nighttime traffic object detection via adaptively integrating event and frame domains

Intelligent perception is crucial in Intelligent Transportation Systems (ITS), with vision cameras as critical components. However, traditional RGB cameras exhibit a significant decline in performance when capturing nighttime traffic scenes, limiting their effectiveness in supporting ITS. In contrast, event cameras possess a high dynamic range (140 dB vs. 60 dB for traditional cameras), enabling them to overcome frame degradation in low-light conditions. Recently, multimodal learning paradigms have made substantial progress in various vision tasks, such as image-text retrieval. Motivated by this progress, we propose an adaptive selection and fusion detection method that leverages both event and RGB frame domains to optimize nighttime traffic object detection jointly. To address the challenge of unbalanced multimodal data fusion, we design a learnable adaptive selection and fusion module. This module performs feature ranking and fusion in the channel dimension, allowing efficient multimodal fusion. Additionally, we construct a novel multi-level feature pyramid network based on multimodal attention fusion. This network extracts potential features to enhance robustness in detecting nighttime traffic objects. Furthermore, we curate a dataset for nighttime traffic scenarios comprising RGB frames and corresponding event streams. Through experiments, we demonstrate that our proposed method outperforms current state-of-the-art techniques in event-based, frame-based, and event and frame fusion methods. This highlights the effectiveness of integrating the event and frame domains in enhancing nighttime traffic object detection.

ESA: External Space Attention Aggregation for Image-Text Retrieval

Due to the large gap between vision and language modalities, effective and efficient image-text retrieval is still an unsolved problem. Recent progress devotes to unilaterally pursuing retrieval accuracy by either entangled image-text interaction or large-scale vision-language pre-training in a brute force way. However, the former often leads to unacceptable retrieval time explosion when deploying on large-scale databases. The latter one heavily relies on the extra corpus to learn better alignment in the feature space while obscuring the contribution of the network architecture. In this work, we aim to investigate a trade-off to balance effectiveness and efficiency. To this end, on the premise of efficient retrieval, we propose the plug-and-play External Space attention Aggregation (ESA) module to enable element-wise fusion of modal features under spatial dimensional attention. Based on flexible spatial awareness, we further propose the Self-Expanding triplet Loss (SEL) to expand the representation space of samples and optimize the alignment of embedding space. The extensive experiments demonstrate the effectiveness of our method on two benchmark datasets. With the identical visual and textual backbones, our single model has outperformed the ensemble modal of similar methods, and our ensemble model can further expand the advantage. Meanwhile, compared with the vision-language pre-training embedding-base method that used 83× image-text pairs than ours, our approach not only surpasses in performance but also accelerates 3× on retrieval time. Codes and pre-trained models are available at https://github.com/KevinLight831/ESA.