国際会議予稿をベースにジャーナル化する際の要件

自分に関係しそうな触覚・VR・HCI分野のジャーナルを調査.
今後も追記予定.

Journal 予稿との差分要件 IF(2019/1)
IEEE Transactions on Haptics 数値の記載なし 1.869
IEEE Transactions on Visualization and Computer Graphics 数値の記載なし 3.078
IEEE Transactions on Multimedia 数値の記載なし 3.977
IEEE Transactions on Human-Machine Systems 数値の記載なし 2.563
IEEE Access 65%以上 3.557
ACM Transactions on Human-Robot Interaction 25%以上
ACM Transactions on Applied Perception 25%以上
ACM Transactions on Multimedia Computing, Communications, and Applications 25%以上
ACM Transactions on Graphics 25%以上
ACM Transactions on Computer-Human Interaction 数値の記載なし
The International Journal of Robotics Research (SAGE Journal) 数値の記載なし
International Journal of Human-Computer Studies (ELSEVIER) 50%以上
Presence (MIT Press) 30%以上
Springer Virtual Reality 100%
Springer Journal on Multimodal User Interfaces 100%
Springer Computer Supported Cooperative Work (CSCW) 30%以上
ICT Frontier 30%以上
Journal of Human-Robot Interaction 50%以上
The Visual Computer 30%以上
Ergonomics (Taylor & Francis) 数値の明記なし

「会議名 duplication」とかでググると投稿規定のページがヒットする.
わかったこととして,ジャーナルごとに差分として必要とされる要件がかなりバラつきがある.
基本的にIEEE系よりACM系の方が数値が緩い.
差分がXX%と明示的に書いてないジャーナルは危険なニオイがする...!

IEEE Transactions on Haptics

情報源

However, papers previously published in conference proceedings, digests, preprints, or records are eligible for consideration provided that the papers have undergone substantial revision and that the author informs the Transactions Assistant and editor-in-chief at the time of submission. A version of the previously published work will need to be sent to our office, and a brief description of the differences will need to be provided

  • 要件
    • substantialなrevisionがあること
    • 投稿時にTransactions Assistantとeditor-in-chiefにrevision箇所を通知すること

IEEE Transactions on Visualization and Computer Graphics

情報源

Papers previously published in conference proceedings, digests, preprints, or records are eligible for consideration provided the editor-in-chief is informed at the time of submission and that they have undergone substantial revision.

  • 要件
    • substantialなrevisionがあること
    • 投稿時にeditor-in-chiefにrevision箇所を通知すること

IEEE Transactions on Multimedia

情報源

Extensions of the Author’s Prior Work. It is acceptable for conference papers to be used as the basis for a more fully developed journal publication. However, authors are required to cite their related prior work, either in the introduction or in a footnote. The papers cannot be identical, and the journal paper should be justifi ed by a clearly identifi able benefi t that its publication offers to the research community beyond the already published conference paper. For example, the journal paper may include additional analysis, novel algorithmic enhancements, added theoretical work, completeness of exposition, extensive experimental validation, etc. The added benefi ts of the journal paper must either be apparent from a reading of the introduction or abstract, or should be clearly and concisely explained in a separate document that accompanies the submission.

  • 要件
    • 引用すること
    • 同一であってはならない
    • ジャーナル化する以下のようなメリットが存在すること
      • 分析の追加
      • アルゴリズムの改善
      • 理論の追加
      • 説明の追加
      • 実験・検証の追加
    • 上記のメリットintroductionかabstract,もしくは1節に分けて明記すること
    • 差分の量は明記されていない

IEEE Transactions on Human-Machine Systems

情報源

If the manuscript or a considerable part of it has been presented, published, or submitted for publication elsewhere, you must inform the Editor. In the cover letter, please explain the contribution of the present work and describe and cite the prior work so that the contribution of the new work is clear. Failure to do so may result in the immediate rejection of your manuscript. Our primary objective is to publish technical material not otherwise available elsewhere.

  • 要件
    • 数値の記載なし

IEEE Access

情報源

IEEE Access accepts expanded versions of an author’s own referenced conference articles, as long as the similarity is less than 35%. If the overlap is higher (effectively duplicate publication) or the source article is not referenced, the article will be returned to the author.

  • 同じ箇所が35%以下であること(=差分が65%以上)

ACM Transactions on Human-Robot Interaction (THRI)

情報源

the paper has been substantially revised (this generally means that at least 25% of the paper is material not previously published; however, this is a somewhat subjective requirement that is left up to each publication to interpret);

  • 要件
    • 25%の差分が必要

ACM Transactions on Applied Perception

情報源

the paper has been substantially revised (this generally means that at least 25% of the paper is material not previously published; however, this is a somewhat subjective requirement that is left up to each publication to interpret); upon submission, the author notifies the EiC(s) or PC(s) that the paper has been previously published or simultaneously submitted; and the published policies of the publications or conferences involved do not prohibit this.

  • 要件
    • 25%の差分が必要

ACM Transactions on Multimedia Computing, Communications, and Applications

情報源

Submissions should contain original material that has not been previously published in a journal, nor is currently under review by another journal. If material in the submission was previously published in a conference paper, the new submission must technically extend the conference paper by at least 25% new material, the submission should clearly cite the prior conference paper, and the authors must upload a letter explaining what has been extended in the new submission. Self-plagiarism is handled similar to all kinds of plagiarism by directly rejecting the article. This encompasses verbatim copying of text passages without marking them as a quote as well as copying of ideas without mentioning the original ressource in the respective text passage. If your paper is a re-submission of a previously rejected submission to TOMM please add a letter-of-changes adressing all comments by the reviewers from the previous submission.

  • 25%の差分が必要

ACM Transactions on Graphics

情報源

TOG allows submitting revised versions of conference or workshop papers; the following guidelines should be considered:

The revision should contribute in a significant way to the value of the presented research - enough that the readers feel that they would direct their students or colleagues to read the journal version. Those additions might include a significant number of additional experiments, data sets, algorithms, or applications of the work. The revision should also reflect more deeply on the limitations. A technical requirement is that at least 25% of the paper are unpublished material. The authors provide the previous version of the paper as "Supplemental File(s) not for review". The authors include a document as "Supplemental File(s) not for review" outlining the differences between the previous and the current submission that describes the significant changes.
Whether the revision is significant is determined by the Associate Editor handling the submission and the Editor-in-Chief. The published policies of the publications or conferences involved must not prohibit the submission to TOG. Note that for papers that have already been published and presented in a conference, there will be no SIGGRAPH/SIGGRAPH Asia presentation.

  • 25%の差分が必要

ACM Transactions on Computer-Human Interaction

情報源

We recognize that occasionally submissions to TOCHI are revised versions of papers that have been considered elsewhere. As long as the paper is not currently being reviewed in any other venue, you are welcome to submit your paper to TOCHI. However, sometimes TOCHI reviewers were the same individuals that reviewed an earlier version of your paper. And in any case, reviewers are a scarce resource, and it is important for the health of our research community to use that resource efficiently and with respect. So, we ask that on a voluntary basis, you include with your submission information about reviews of prior versions of the current submission that you think will help the reviewers make an informed and efficient decision. For example, this could take the form of including the original reviews and venue name. Or, you could summarize what you think were the key concerns and how you refined the paper in response to those concerns.

  • 差分の数値は明記されていない

The International Journal of Robotics Research

情報源

Subject to the journal's policy, manuscripts based on papers that have been presented at conferences may be considered for publication as long as they have not been published and provided that you still retain the rights to the manuscript. The journal editor may review whether the version of your article considered for publication is materially different from the work you presented at a conference and/or whether publication in the journal will enable your article to reach an audience that the conference paper did not previously reach. Prior publication of an abstract or poster presented at a conference will generally not impact the manuscript's eligibility for publication.

  • 要件
    • 数値は明記されていない.

International Journal of Human-Computer Studies

情報源

The Editors in Chief will consider papers previously published in conference proceedings, provided that i) the authors inform the editorial office about this in the submission cover letter, ii) highlight the novel parts in the new submission, and iii) the journal version includes significant extensions and enhancements to the conference version. As a rough guideline, at least 50 per cent of the submitted paper should be original and not published elsewhere.

  • 要件
    • editorial officeにcover letter内で予稿について知らせること
    • 論文と会議予稿の差分をハイライトすること
    • 少なくとも50%の差分が存在すること

Presence

情報源

When a manuscript is submitted to Presence, it is assumed that it has not been published elsewhere, nor is it being considered for publication elsewhere, in a peer reviewed journal or a commercially available proceedings of a meeting, independent of the language of that publication. This constraint shall remain in effect until the final Presencereview process has been completed and the manuscript has been accepted or rejected by Presence. When there is any doubt, authors should include copies of the previously published or submitted manuscript and a detailed statement of differences for consideration by the reviewers and editors along with the Presence submission. To be considered an original work, extended journal manuscripts based on prior publications must contain significant amplification or clarification of the original material (a minimum of 30% new material when compared with the prior publication). Violation of this requirement will automatically result in rejection of the submission to Presence.

  • 差分が30%以上存在すること

Springer Virtual Reality

情報源

Submission of a manuscript implies: that the work described has not been published before; that it is not under consideration for publication anywhere else; that its publication has been approved by all co-authors, if any, as well as by the responsible authorities – tacitly or explicitly – at the institute where the work has been carried out. The publisher will not be held legally responsible should there be any claims for compensation.

  • 完全に新規であること

Springer Journal on Multimodal User Interfaces

情報源

Submission of a manuscript implies: that the work described has not been published before; that it is not under consideration for publication anywhere else; that its publication has been approved by all co-authors, if any, as well as by the responsible authorities – tacitly or explicitly – at the institute where the work has been carried out. The publisher will not be held legally responsible should there be any claims for compensation.

  • 完全に新規であること

Springer Computer Supported Cooperative Work (CSCW)

情報源

The CSCW Journal only publishes original research. As for work previously reported in conference papers, there must be sufficient difference between the conference paper and the version submitted to the Journal for the two works to be seen as separate works. As a rule of thumb, the journal version must be updated by 30% or more (e.g., extended methodology, more research results, new figures and or tables, etc.) and, hence, have a new title.

  • 要件
    • 30%以上の差分があること

ICT Frontier

情報源

  • 要件
    • substantial amount of original contentがあること
      具体的には
      • 新しい素材(実験,データ)
      • 新たな議論や結論につながる,古いデータに対する新たら解析
      • 目安として少なくとも30%のコンテンツが新しいこと
    • 会議予稿のcopyrightを著者が持ってない場合は内容再利用の許可を得ること
    • Acknowledgementsで会議予稿を引用するか,参考文献の頁で引用すること

Journal of Human-Robot Interaction

情報源

Research Reports. Research reports (up to 4,000 words) are expected to present new research findings and will be favored if they present innovations in approach or method. Research reports are likely to be published more quickly than research articles. Overlap with a conference paper may be no more than fifty percent. We especially encourage reports, which count as “articles” for archiving purposes just the same as regular articles. Recent research suggests that shorter papers are downloaded, printed, read, and cited more than longer articles.

  • 要件
    • 差分が50%以上あること

The Visual Computer

情報源

Authors please note that any submissions to the journal that have previously been published in conference proceedings MUST contain at least 30% new content with a new case study, new title, new images and additional results. Any submissions that do not adhere to the guidelines above will be considered as an attempt at Duplicate Publication and will be rejected by the journal, with the authors facing a potential ban on any future submissions.

  • 要件
    • 差分が30%以上あること

Ergonomics (Taylor & Francis)

情報源

  • 要件
    • 数値の明記なし

arXiv投稿メモ

画像サイズは小さく(1MB以下)にする必要あり

自分はpdfにして,そのあとepsに戻すスクリプトで一括変換している.

for file in `\find ./figures -maxdepth 1 -type f`; do
    echo $file
    epstopdf $file
    pdftops -eps ${file/eps/pdf}
done
.bibファイルとともに同名の.bblファイルが必要.

ないときはbibファイルからbibtexで作る. .bibファイル,.bblファイルともにtexファイルと同名にする.

クレジットを入れる

プレプリントを投稿する際,出版社によってはクレジットを1ページ目のどこかに載せることを要件としていることがある. ヘッダに載せるにはfancy.styを使って\maketitle の後あたりに以下のように載せれば良い.

\usepackage{fancyhdr}

\thispagestyle{fancy}
\fancyhf{}
\chead{XXXXXXXXX}

\renewcommand{\headrulewidth}{0pt} % 罫線を消す

epsファイルサイズを減らす

arXiv対策等でepsファイルサイズをとにかく減らしたい時がある.
そんな時はpdf化して再度eps化する.

epstopdf filename.eps
pdftops -eps filename.pdf

もしno display font for 'helvetica-bold'等のエラーが出たら, xpdfをインストールし直す. 私のmac環境ではbrewだと治らなかった. マスクメロンの Xpdf.app を使用する方法でxpdfをインストールし直すと治った.

Food101データセットを触ってみる

Food101という101クラスの料理画像がラベルとセットになっているデータセットがある。 こちらを触ってみる。

クラスごとに250枚のテスト画像と750枚の訓練画像があり、全部で101000枚画像が用意されている。 オリジナルだと画像の最大サイズは同じ(512pixel)にスケールされているものの、もう片方の辺はバラバラのようなので、 kaggleのここを参考にサイズが揃ったデータを取得し可視化だけしてみる。

def visualize_data():
    base_path = "./dataset"
    train_h5_path = os.path.join(base_path, "food_c101_n10099_r64x64x3.h5")
    test_h5_path = os.path.join(base_path, "food_test_c101_n1000_r64x64x3.h5")
    print(os.path.exists(train_h5_path))
    print(os.path.exists(test_h5_path))

    with h5py.File(train_h5_path, 'r') as n_file:
        total_imgs = n_file['images'].shape[0]
        read_idxs = slice(0, total_imgs)

        img_data = n_file['images'][read_idxs]
        img_label = n_file['category'].value[read_idxs]
        label_names = [x.decode() for x in n_file['category_names'].value]

    fig, m_ax = plt.subplots(5, 5, figsize = (12, 12))
    for c_ax, c_label, c_img in zip(m_ax.flatten(), img_label, img_data):
        c_ax.imshow(c_img)
        c_ax.axis('off')
        c_ax.set_title(label_names[np.argmax(c_label)])

f:id:yusuke_ujitoko:20190112122433p:plain