We encourage all authors to read these submission policies carefully before preparing their manuscripts. Following these guidelines will help ensure a smooth review process and fair evaluation of all submissions. Authors are responsible for ensuring that their papers comply with the requirements described below.

ICPRv 2027 welcomes two categories of technical papers, i.e., Long Papers and Short Papers. Authors should select the category that best matches the scope of their contribution. Both categories will undergo double blind peer review, will be published in the conference proceedings, and will be presented as part of the technical program.

• Long Papers: up to 15 pages in LNCS format, excluding references.
• Short Papers: up to 8 pages in LNCS format, excluding references.

Long and Short Papers are intended for different types of contributions. Long Papers are meant to present more comprehensive contributions, while Short Papers are intended for focused contributions that can be presented effectively in a concise format. Both categories are held to the same standards of scientific quality and will be evaluated according to the expectations of their respective category. Additional details on the two paper categories are given below:

• Long Papers (15 pages + references) should present high-quality original research that makes a substantial contribution to pattern recognition or a closely related field. Contributions may include new methods, important insights or findings, significant theoretical developments, or innovative applications supported by clear methodological advances. Authors should position their work within the relevant literature and explain clearly how it differs from and improves upon existing approaches. Research procedures, technical methods, and evaluation protocols should be described in sufficient detail to support careful assessment and, where appropriate, reproducibility. The results should be presented clearly and supported by convincing experimental, analytical, or theoretical evidence. Authors should also discuss the significance of their findings, the limitations of the work, and its potential impact on the pattern recognition community and related fields.
• Short Papers (8 pages  + references) provide a venue for concise, focused contributions that can be communicated effectively within a shorter format. Their scope is intentionally narrower than that of Long Papers, while meeting the same standards of scientific quality, originality, and technical rigor. Examples include focused methodological advances, novel applications, benchmark or dataset papers, empirical analyses, reproducibility studies, negative results, or emerging research directions of interest to the pattern recognition community. Short Papers should present complete and self-contained contributions rather than preliminary or incomplete versions of Long Papers. ICPRv 2027 particularly welcomes the following types of Short Papers :

• Blue Sky Papers, presenting bold, unconventional, or forward-looking ideas that have the potential to open up new research directions. These papers should clearly articulate the motivation and potential impact of the proposed ideas. While the level of experimental validation may differ from that of a comprehensive Long Paper, the contribution should be technically sound and supported by appropriate evidence.
• Application Papers, describing innovative applications of pattern recognition techniques to new problem domains or real-world challenges. The primary contribution may lie in the originality of the application, the insights gained, or the impact of the work, rather than in methodological novelty alone.
• Resource Papers, introducing valuable resources for the community, such as datasets, benchmarks, software, annotation protocols, evaluation methodologies, or reproducibility studies that are expected to facilitate future research.

ICPRv 2027 will use the Microsoft Conference Management Toolkit (CMT) to manage paper submissions and the double-blind review process. Authors should ensure that their manuscripts are properly anonymized and do not disclose the identity of the authors or their institutions.

To facilitate accurate reviewer and Area Chair assignment, ICPRv 2027 will use the Toronto Paper Matching System (TPMS). Information provided during submission, including the paper title, abstract, keywords, and manuscript content, will be processed by these systems to identify the most appropriate reviewers. By submitting a paper to ICPRv 2027, authors agree to the use of CMT and TPMS as part of the conference review process. Submitted manuscripts will be treated as confidential and will only be accessible to the Program Chairs, Area Chairs, and reviewers assigned to the paper.

Detailed instructions on preparing and submitting a manuscript through CMT, together with the LNCS paper templates, are provided on the Paper Submission page.

The main conference submission must consist of a single manuscript in PDF format with a maximum file size of 50 MB. The manuscript must be complete and self-contained, containing all information necessary for reviewers to evaluate the scientific contribution.

Authors may optionally submit supplementary material to accompany the manuscript. The supplementary material must be uploaded as a single PDF, RAR or ZIP file with a maximum total size of 250 MB  Supplementary material may include additional figures or tables, extended proofs, detailed experimental results, videos, or other supporting content that complements the main paper. Supplementary material must comply with the same anonymization requirements as the main manuscript.

The supplementary material is optional and should not be required to understand or evaluate the main contribution of the paper. Reviewers are not expected to examine the supplementary material but may choose to if they wish . Acceptance decisions will be based primarily on the main manuscript.

For accepted papers, the supplementary material will not be included in the conference proceedings but will be made publicly available through SpringerLink together with the published paper.

The author list is considered final upon the submission deadline. After this deadline, no authors may be added to or removed from a submission. Requests to change the order of authors or to correct author information will be considered only in exceptional circumstances and at the discretion of the Program Chairs. In any case, written approval from all co-authors as well as the department chair of the corresponding author is needed for the request to be considered.

Submitted manuscripts and supplementary material should be self-contained. Authors must not include external links (e.g., webpages, videos, or online demonstrations) that expand the content of the submission or otherwise circumvent the page limits, submission deadline, or review process. Links to publicly available resources that are not part of the scientific contribution (e.g., references to existing datasets or published software) are permitted provided they do not compromise anonymity. Similarly, links to anonymous code repositories are allowed. We encourage authors to upload supporting material as part of the supplementary material.

ICPRv 2027 uses a double-blind review process. Authors do not know the identity of the reviewers or Area Chairs assigned to their paper, and reviewers and Area Chairs should not be able to identify the authors from the submitted materials.

Authors are responsible for ensuring that their submissions are properly anonymized. Manuscripts, supplementary material, and any other submitted files must not contain information that directly or indirectly reveals the identity or affiliation of the authors. This includes, but is not limited to, acknowledgements, funding information, institutional identifiers, author names in figures or videos, metadata, and links to websites or repositories that disclose the authors' identity. References to the authors' own previously published work should be written in the third person in the same manner as any other citation. Authors should not refer to their previous work using phrases such as "our previous work" or "in our earlier paper" if doing so would compromise anonymity. If authors wish to cite unpublished work that is under concurrent review elsewhere, they should provide an anonymized version of that work as supplementary material, cite the anonymized manuscript in the paper, and clearly explain how the ICPRv 2027 submission differs from the concurrent submission. Authors should remove identifying metadata from submitted PDF files whenever possible.

Failure to comply with the double-blind review requirements may result in the paper being rejected without review.

Plagiarism is the presentation of another person's ideas, text, results, figures, or other intellectual contributions without appropriate acknowledgment. ICPRv 2027 expects all submissions to represent the authors' own original work and to properly cite the work of others. All submissions will be screened for plagiarism using appropriate software tools and other means. Suspected cases of plagiarism or other forms of publication misconduct will be investigated by the Program Chairs and handled in accordance with the policies of Springer and the International Association for Pattern Recognition (IAPR), as appropriate. Authors are reminded that plagiarism, including self-plagiarism and the unattributed reuse of previously published material, is considered a serious breach of scientific ethics and may result in rejection of the submission and further actions in accordance with the applicable publication policies.

ICPRv 2027 aims to publish original research contributions that have not been previously published and to avoid duplicating the effort of reviewers. By submitting a manuscript to ICPRv 2027, the authors confirm that the work has not been previously published or accepted for publication in substantially similar form in any peer-reviewed venue, including journals, conferences, workshops, or other archival publications. Preprints and technical reports that have not undergone peer review (e.g., arXiv, institutional repositories, or personal webpages) are not considered prior publications (i.e. work that has been published or accepted for publication, regardless of its length and publication format) and do not violate this policy. Authors may cite such material where appropriate, but are not required to do so.

Authors further confirm that no substantially similar manuscript has been or will be submitted to another peer-reviewed venue while the ICPRv 2027 submission is under review. Submissions are considered substantially similar if they share a significant portion of their technical content, results, or conclusions. Violations of this policy may result in rejection of the submission and notification of the other venue. For the purposes of this policy, a prior publication is any work that has undergone peer review and has been accepted for publication, regardless of its length, publication format and whether the publication appears in formal proceedings.

If a paper is accepted, at least one author must register for ICPRv 2027 and present the paper during the conference. Presentations will be delivered in the virtual conference format according to the conference schedule. Failure to present an accepted paper may result in its removal from the conference proceedings and associated publication.

The quality of ICPRv 2027 depends on the active participation of the pattern recognition community in the peer-review process. Authors who submit papers to the conference are therefore expected to contribute to the review process by serving as reviewers if invited. To help identify qualified reviewers, the corresponding author will be asked to nominate one or more researchers from the authors list with appropriate expertise in the subject area of the submission. These nominations will be used only as one source of information when selecting reviewers and do not guarantee that the nominated individuals will be invited.

All authors must be registered in the Microsoft Conference Management Toolkit (CMT) using the same email address that is provided in the paper submission. Whenever possible, authors are strongly encouraged to use their institutional email address.

To ensure the integrity of the review process, authors are responsible for accurately declaring all conflicts of interest in CMT. This includes specifying all relevant email domains and any other conflicts required by the submission system. Failure to disclose known conflicts may compromise the review process and could result in the rejection of the submission.

To preserve the integrity of the review process, authors should not issue press releases or otherwise seek media coverage of submitted work while it is under review for ICPRv 2027. Public communication of research through preprints, institutional repositories, or personal webpages is permitted in accordance with the conference's policies on prior publication and double-blind review. Papers may be publicized through the media only after they have been officially accepted for publication. Failure to comply with this policy may result in the paper being withdrawn from the conference and its proceedings.

ICPRv 2027 permits authors to post preprints of their manuscripts on non-peer-reviewed public repositories, such as arXiv, institutional repositories, or personal webpages, before and during the review process. Such preprints are not considered prior publications and do not violate the conference's dual submission policy.

Authors may also maintain public repositories (e.g., GitHub) or project webpages related to their research. However, to preserve the integrity of the double-blind review process, submitted manuscripts and supplementary material must not contain links or references that reveal the identity of the authors. Authors should therefore avoid citing their own public repositories or project webpages if doing so would compromise anonymity.

As part of good scholarly practice, authors are expected to cite relevant prior work, including preprints that have influenced their research. However, authors are not expected to compare against or discuss preprints that appeared shortly before the submission deadline if doing so would impose an unreasonable burden.

The personal information provided during the submission and registration process will be used solely for the administration of ICPRv 2027 and related conference activities. This may include communication with authors regarding their submissions, the review process, registration, and future ICPR or IAPR events.

Author contact information may be shared with trusted third-party service providers acting on behalf of ICPRv 2027 or the International Association for Pattern Recognition (IAPR) for the purposes described above. Personal information will not be shared with third parties for commercial or marketing purposes unrelated to the conference.

Authors whose work involves personal data, human participants, personally identifiable information, or potentially sensitive or offensive content are expected to ensure that the data were collected, processed, and used in accordance with applicable ethical guidelines and legal requirements.

Where required by the authors' institution or applicable regulations, studies involving human participants or personal data should have received approval from an Institutional Review Board (IRB) or an equivalent ethics committee. If such approval was obtained, authors should state this in the manuscript. If approval was not required, was pending, or could not be obtained, authors should explain the circumstances, describe how the data were collected, and indicate whether informed consent was obtained or justify why it was not.

This information may be included either in the main paper or in the supplementary material. Authors using existing public datasets are encouraged to consider how the data were collected and whether appropriate consent and ethical approval were obtained.

ICPRv 2027 encourages authors to support open and reproducible research by making the research artifacts associated with their work publicly available whenever possible. Such artifacts may include source code, datasets, trained models, annotations, configuration files, or other materials needed to reproduce the reported results. To preserve the integrity of the double-blind review process, authors should not include links to these resources in the submitted manuscript or supplementary material if doing so would reveal their identity. Instead, authors are encouraged to indicate that the corresponding artifacts will be made publicly available upon acceptance.

If a paper claims the release of a dataset, software package, or other research artifact as one of its primary contributions, the corresponding artifact is expected to be publicly available no later than the camera-ready deadline. The use of private or otherwise restricted datasets or resources is acceptable; however, such resources should not be presented as publicly available contributions if they cannot be shared with the research community.

ICPRv 2027 will continue the Reproducible Research in Pattern Recognition (RRPR) initiative. Accepted papers may be eligible to receive an RRPR Badge following a separate evaluation by the Reproducibility Chairs. 

ICPRv 2027 welcomes the responsible use of tools, including Large Language Models (LLMs), to support research and manuscript preparation. However, authors remain fully responsible for the originality, accuracy, and integrity of all submitted material, regardless of the tools used in its preparation.

The use of an LLM or other AI tool must be disclosed if it forms part of the research methodology or contributes to the scientific results reported in the paper. Examples include, but are not limited to, data generation or filtering, annotation, experimental design, code generation where it materially affects the proposed method, theorem proving, or any other aspect that is necessary for understanding, reproducing, or evaluating the research. Such use should be described in sufficient detail to ensure transparency and reproducibility.

The use of LLMs solely to improve the language, grammar, style, or readability of the manuscript does not need to be disclosed. Regardless of how the manuscript was prepared, authors are responsible for ensuring that the paper is original, accurately represents the reported research, properly acknowledges the work of others, and complies with the conference policies on plagiarism and research integrity.

Authors must not include hidden text, hidden instructions, or other content intended to manipulate reviewers or automated systems (e.g., prompt injection). Such behavior constitutes scientific misconduct and may result in rejection without review and further action by the Program Chairs.

Questions regarding the submission policies, paper preparation, or the review process should be directed to the Program Chairs at icpr27pc@iapr.org. To ensure consistent communication, authors are kindly requested to contact the Program Chairs directly rather than members of the Program Committee or Organizing Committee.