{
  "id": 687043,
  "title": "Request for Dataset Access – Academic Research",
  "url": "/competitions/recodai-luc-scientific-image-forgery-detection/discussion/687043",
  "author_name": "",
  "post_date": "2026-04-01T22:49:47.906088100Z",
  "votes": null,
  "comment_count": 1,
  "views": 0,
  "content": "<p>Dear Dr. Cardenuto <a href=\"https://www.kaggle.com/joophillipecardenuto\" target=\"_blank\">@joophillipecardenuto</a> </p>\n<p>I hope you are doing well. I am pursuing MSc in Integrated Machine Learning Systems at University College London (UCL). As part of my academic research, I am required to select a Kaggle competition and develop a hypothesis-driven research solution, written up as a conference-style paper.</p>\n<p>Your Scientific Image Forgery Detection competition is an excellent fit for my project. I am particularly interested in comparing segmentation architectures for detecting copy-move forgeries in biomedical images — a topic that aligns well with both the competition's goals and my module's requirements.</p>\n<p>Unfortunately, I am unable to join the competition through the standard route as the entry deadline appears to have passed, which prevents me from accessing the dataset. Would it be possible to either:</p>\n<ul>\n<li><p>Extend or reopen entry so I can download the data via Kaggle, or</p></li>\n<li><p>Provide an alternative means of accessing the dataset for purely academic and educational purposes?</p></li>\n</ul>\n<p>I would fully comply with any usage restrictions and will appropriately cite your work in my report.</p>\n<p>Thank you very much for organising such a well-structured and impactful competition. I look forward to hearing from you.</p>\n<p>Kind regards,\nKamal Jain</p>",
  "messages": [
    {
      "id": "3433648",
      "postDate": "04/01/2026 22:49:47",
      "content": "<p>Dear Dr. Cardenuto <a href=\"https://www.kaggle.com/joophillipecardenuto\" target=\"_blank\">@joophillipecardenuto</a> </p>\n<p>I hope you are doing well. I am pursuing MSc in Integrated Machine Learning Systems at University College London (UCL). As part of my academic research, I am required to select a Kaggle competition and develop a hypothesis-driven research solution, written up as a conference-style paper.</p>\n<p>Your Scientific Image Forgery Detection competition is an excellent fit for my project. I am particularly interested in comparing segmentation architectures for detecting copy-move forgeries in biomedical images — a topic that aligns well with both the competition's goals and my module's requirements.</p>\n<p>Unfortunately, I am unable to join the competition through the standard route as the entry deadline appears to have passed, which prevents me from accessing the dataset. Would it be possible to either:</p>\n<ul>\n<li><p>Extend or reopen entry so I can download the data via Kaggle, or</p></li>\n<li><p>Provide an alternative means of accessing the dataset for purely academic and educational purposes?</p></li>\n</ul>\n<p>I would fully comply with any usage restrictions and will appropriately cite your work in my report.</p>\n<p>Thank you very much for organising such a well-structured and impactful competition. I look forward to hearing from you.</p>\n<p>Kind regards,\nKamal Jain</p>",
      "rawMarkdown": "Dear Dr. Cardenuto @joophillipecardenuto \n\nI hope you are doing well. I am pursuing MSc in Integrated Machine Learning Systems at University College London (UCL). As part of my academic research, I am required to select a Kaggle competition and develop a hypothesis-driven research solution, written up as a conference-style paper.\n\nYour Scientific Image Forgery Detection competition is an excellent fit for my project. I am particularly interested in comparing segmentation architectures for detecting copy-move forgeries in biomedical images — a topic that aligns well with both the competition's goals and my module's requirements.\n\nUnfortunately, I am unable to join the competition through the standard route as the entry deadline appears to have passed, which prevents me from accessing the dataset. Would it be possible to either:\n\n- Extend or reopen entry so I can download the data via Kaggle, or\n\n- Provide an alternative means of accessing the dataset for purely academic and educational purposes?\n\nI would fully comply with any usage restrictions and will appropriately cite your work in my report.\n\nThank you very much for organising such a well-structured and impactful competition. I look forward to hearing from you.\n\nKind regards,\nKamal Jain",
      "votes": null
    },
    {
      "id": "3434845",
      "postDate": "04/03/2026 10:13:01",
      "content": "<p>Hi Kamal,</p>\n<p>We are unfortunately unable to allow new entries at this stage. Since the competition involves real-world cases, we must keep the current solutions \"frozen\" to ensure there is no data leakage. This ensures that the models are tested against data they haven't seen before, maintaining the integrity of the evaluation.</p>\n<p>Once the competition officially closes, we will release the full testset. We are also planning to transform the data into a Kaggle Benchmark to facilitate future research and comparisons.</p>\n<p>Kind Regards,\nJoão</p>",
      "rawMarkdown": "Hi Kamal,\n\nWe are unfortunately unable to allow new entries at this stage. Since the competition involves real-world cases, we must keep the current solutions \"frozen\" to ensure there is no data leakage. This ensures that the models are tested against data they haven't seen before, maintaining the integrity of the evaluation.\n\nOnce the competition officially closes, we will release the full testset. We are also planning to transform the data into a Kaggle Benchmark to facilitate future research and comparisons.\n\nKind Regards,\nJoão",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 3434845,
      "author_name": "joophillipecardenuto",
      "author_url": "",
      "post_date": "04/03/2026 10:13:01",
      "content": "<p>Hi Kamal,</p>\n<p>We are unfortunately unable to allow new entries at this stage. Since the competition involves real-world cases, we must keep the current solutions \"frozen\" to ensure there is no data leakage. This ensures that the models are tested against data they haven't seen before, maintaining the integrity of the evaluation.</p>\n<p>Once the competition officially closes, we will release the full testset. We are also planning to transform the data into a Kaggle Benchmark to facilitate future research and comparisons.</p>\n<p>Kind Regards,\nJoão</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "3433648": "Dear Dr. Cardenuto @joophillipecardenuto \n\nI hope you are doing well. I am pursuing MSc in Integrated Machine Learning Systems at University College London (UCL). As part of my academic research, I am required to select a Kaggle competition and develop a hypothesis-driven research solution, written up as a conference-style paper.\n\nYour Scientific Image Forgery Detection competition is an excellent fit for my project. I am particularly interested in comparing segmentation architectures for detecting copy-move forgeries in biomedical images — a topic that aligns well with both the competition's goals and my module's requirements.\n\nUnfortunately, I am unable to join the competition through the standard route as the entry deadline appears to have passed, which prevents me from accessing the dataset. Would it be possible to either:\n\n- Extend or reopen entry so I can download the data via Kaggle, or\n\n- Provide an alternative means of accessing the dataset for purely academic and educational purposes?\n\nI would fully comply with any usage restrictions and will appropriately cite your work in my report.\n\nThank you very much for organising such a well-structured and impactful competition. I look forward to hearing from you.\n\nKind regards,\nKamal Jain",
    "3434845": "Hi Kamal,\n\nWe are unfortunately unable to allow new entries at this stage. Since the competition involves real-world cases, we must keep the current solutions \"frozen\" to ensure there is no data leakage. This ensures that the models are tested against data they haven't seen before, maintaining the integrity of the evaluation.\n\nOnce the competition officially closes, we will release the full testset. We are also planning to transform the data into a Kaggle Benchmark to facilitate future research and comparisons.\n\nKind Regards,\nJoão"
  },
  "source": "meta"
}