Science AI Journal
Computer Science · Research topic

Open research questions in Image Retrieval and Classification Techniques

77 gap statements mined from Image Retrieval and Classification Techniques papers in our 4.5M-paper local library, which holds 641 papers on the topic — drawn mostly from each paper's own stated research gap, future-work, challenge and limitation notes, and its abstract. The ones listed below are a selection still marked open; each names the study that raised it, with a DOI link where the paper has one.

Choosing where to publish on Image Retrieval and Classification Techniques? See the ranked Computer Science journals.

What the literature leaves open

  • The lack of a simple and easy-to-use system for guests to retrieve their photos. The need for a system that can achieve high accuracy in matching guests with their photos.

    AI-Driven Guest Identification and Photo Retrieval System · 2026 · DOI

  • Cross-validation strategies, different data split ratios, and their impact on BGP-Model robustness across batik motif categories are not explored.

    Batik Motif Recognition Using the BGP-Model: A Hybrid GLCM-PCA Approach with Machine Learning Classifiers · 2026 · DOI

  • The lack of paired data can make it challenging to ensure high-quality translations and maintain semantic coherence in the output images. Obtaining paired datasets can be impractical for many applications, as it requires manual annotation of images.

    Unpaired image-to-image translation with content preserving perspective: a review · 2026 · DOI

  • The study highlights the need for a systematic investigation of the effectiveness of supervised and self-supervised backbones as feature extractors for artwork classification.

    Harnessing Self-Supervised Features for Art Classification · 2026

  • Future studies are recommended to incorporate additional metrics, such as Structural Similarity Index (SSIM) and Peak Signal-to-Noise Ratio (PSNR), to provide a more comprehensive evaluation of image qual.

    Analysis of Digital Image Quality Between Conventional Design Outputs and Generative Artificial Intelligence Images Using Digital Image Processing Methods · 2026 · DOI

  • None of these studies evaluate multimodal complaint systems on their ability to handle sarcasm, irony, or non-literal expressions in citizen reports—challenges documented in multimodal stance detection but never tested in civic complaint contexts, where such expressions may obscure genuine infrastructure issues.

    Multimodal detection of urban improvement indicators in public visual-text content using deep learning · 2026 · DOI

  • The requirement for extensive manual labeling and significant computing power is a major barrier to the implementation of deep learning approaches. The lack of user-friendly and accessible tools for non-AI specialists is a significant challenge.

    IAMAP: Unlocking Deep Learning in QGIS for non-coders and limited computing resources · 2026 · DOI

  • Existing zero-shot semantic segmentation frameworks are limited by their reliance on manual annotation and closed-world assumptions. The large number of classes and complex distributions make comprehensive annotation virtually impossible.

    Knowledge Graph Enhanced for Zero-Shot Semantic Segmentation in Remote Sensing Imagery · 2026 · DOI

  • The limited scalability and insufficient real-time adaptability of most existing intelligent multimodal systems.

    Designing Intelligent Multimodal Assistants for Digital Humanities: A Comparative Study of Models, Modalities, and Domains · 2026 · DOI

  • The rapid growth of video content has made efficient video retrieval a critical challenge. Traditional approaches rely heavily on manual tagging and metadata generation, which are time-consuming, inconsistent, and not scalable. The lack of efficient video retrieval systems that can handle large-scale video data.

    Semantic Video Discovery Using Deep Feature Fusion And Automated Metadata Generation · 2026 · DOI

  • There is a need for a data model to add structured data to natural history specimen images on Wikimedia Commons. The existing data model does not support a wider range of Wikimedia Commons queries.

    Wikimedia Commons data model for Natural History specimen images · 2026 · DOI

  • The development of MLLMs in the RS domain is still in the infant stage. There is a significant domain gap between natural and remote sensing images. Current MLLMs have limitations in terms of their ability to follow instructions and perform inference for RS downstream tasks.

    EarthGPT: A Universal Multimodal Large Language Model for Multisensor Image Comprehension in Remote Sensing Domain · 2024 · DOI

  • Future studies can use the identified dimensions to develop more effective visual communication strategies. The study's results can be applied to various fields such as advertising, education, and art. Further research can be conducted to examine the universality of the identified dimensions across different cultures and populations.

    Identification of visual dimensions in photographs using multidimensional scaling techniques · 1984 · DOI

  • The lack of research on interpretation of visuals and corresponding cognitive processes. The necessity for combining appropriate methods from two seemingly incompatible research paradigms.

    Identification of visual dimensions in photographs using multidimensional scaling techniques · 1984 · DOI

  • The paper identifies a gap in the literature, which has traditionally focused on central place theory, neglecting the importance of gateways. The study highlights the need to consider both central place and gateway developments in understanding the rise of large centers.

    GATEWAYS: SLOW RECOGNITION BUT IRRESISTIBLE RISE · 1983 · DOI

  • The challenge of teaching reading and language skills to mentally retarded students who may have limited cognitive abilities. The challenge of finding available material at the students' skill level. The challenge of teaching students who may have additional disabilities, such as deafness.

    My Best Friends Are Limited-But Who's Perfect · 1979 · DOI

  • The threshold θ for cosine similarity matching was set conservatively to prioritize precision over recall, but no systematic analysis of threshold selection or optimization methodology is provided.

    AI-Driven Guest Identification and Photo Retrieval System · 2026 · DOI

  • Misclassification patterns are noted (e.g., Parangsloboh misclassified as Parangklitih) but root cause analysis is absent. Systematic investigation of which GLCM texture features fail to discriminate between visually similar batik motifs would inform feature engineering improvements.

    Batik Motif Recognition Using the BGP-Model: A Hybrid GLCM-PCA Approach with Machine Learning Classifiers · 2026 · DOI

  • There are many security, performance, scalability, and architectural design issues when it comes to implementing such APIs in web apps. Most of the applications designed for academic or research purposes are very much experimental minded.

    Unsplash FindAWall: Secure, Scalable and Responsive Image-Search Platform · 2026 · DOI

  • The dataset was small and imbalanced. Some monuments had fewer than 10 samples. The system requires further development for multilingual support and AR/VR support.

    Deep Learning Framework for Indian Heritage Site Classification and Virtual Tour Generation using Transfer Learning · 2026 · DOI

  • The study aims to address the question of whether multi-image presentation improves concept acquisition. The research seeks to determine whether redundancy and an opportunity to view nonexamples would be helpful to seventh-grade subjects.

    Implications of multi-image for concept acquisition · 1979 · DOI

  • Multilingual support for the system. AR/VR support for the system. Implementation as an offline mobile/web application.

    Deep Learning Framework for Indian Heritage Site Classification and Virtual Tour Generation using Transfer Learning · 2026 · DOI

  • The proposed model consists of two novel modules: the staircase structure and the iterative mixed database training (IMDT) strategy. Jiang, “Unified quality assessment of in-the-wild videos with mixed datasets training,” International Journal of Computer Vision, vol.

    Blind Quality Assessment for in-the-Wild Images via Hierarchical Feature Fusion and Iterative Mixed Database Training · 2023 · DOI

  • The paper suggests that future research could explore the application of principles from philosophy of language to other varieties of images - The author notes that further study is needed to fully understand the factors that determine the identification of an image…

    How to do things with images · 1980 · DOI

  • The paper identifies a gap in the field of image analysis, specifically in the application of principles from philosophy of language - The author notes that previous studies have not adequately addressed the analysis of images using principles from philosophy…

    How to do things with images · 1980 · DOI

Most-cited papers in Image Retrieval and Classification Techniques

Most recent work

Find a gap in your own Image Retrieval and Classification Techniques sub-topic

This page shows what the Image Retrieval and Classification Techniques literature already flags as unresolved. To narrow it to your specific question, search the Research Gap Finder: the search is free with a free account and lists the papers closest to your topic first. Unlocking that topic (50 credits, charged once) fills the comparison table from our 4.5M-paper local library and writes the gaps from its rows.

Open the Research Gap Finder →

Related topics in Computer Science

Online Learning and Analytics
887 open questions
AI in Service Interactions
805 open questions
Quantum Computing Algorithms and Architecture
632 open questions
Virtual Reality Applications and Impacts
610 open questions
Teaching and Learning Programming
576 open questions
Digital literacy in education
522 open questions
Computational Drug Discovery Methods
505 open questions
Topic Modeling
482 open questions
Open questions raised by papers listed on this page include: (1) The lack of a simple and easy-to-use system for guests to retrieve their photos (“AI-Driven Guest Identification and Photo Retrieval System”, 2026); (2) The lack of paired data can make it challenging to ensure high-quality translations and maintain semantic coherence in the output images (“Unpaired image-to-image translation with content preserving perspective: a…”, 2026); (3) The lack of user-friendly and accessible tools for non-AI specialists is a significant challenge (“IAMAP: Unlocking Deep Learning in QGIS for non-coders and limited computing…”, 2026); (4) The lack of efficient video retrieval systems that can handle large-scale video data (“Semantic Video Discovery Using Deep Feature Fusion And Automated Metadata…”, 2026); (5) There is a need for a data model to add structured data to natural history specimen images on Wikimedia Commons (“Wikimedia Commons data model for Natural History specimen images”, 2026). 77 gap statements have been mined from Image Retrieval and Classification Techniques papers in our 4.5M-paper local library, which holds 641 papers on the topic; the gaps come from whichever of those papers state one. They are mostly the research gaps the authors state and the papers' abstracts, plus future-work, limitations and challenges passages. The ones listed below are a selection still marked open; each names the study that raised it, with a DOI link where the paper has one, so you can read the original claim in context.

Tools for your next paper

Gaps & Literature SearchFind open questions in any field — 120,000+ open gaps mined from our own library.Pre-CheckIs your paper ready? Acceptance odds for top, mid and open/emerging journals, in seconds.Journal FinderA ranked shortlist of target journals from a 17,500-venue index.Duplicate Publication CheckerCheck for prior publication across 7 sources and our local library.Citation GeneratorFreeA DOI or title → APA, MLA, Vancouver, Chicago, BibTeX, RIS. No signup.Graphical Abstract MakerFreeTurn your findings into a colorblind-safe graphical abstract.

How Pre-Check worksHow AI Review works

Compare the category — Honest roundups of the AI research tools, ours listed alongside the alternatives.

Command palette

Jump anywhere, run any action.