{
  "id": 5881,
  "title": "If you see something, say something.",
  "url": "/competitions/dogs-vs-cats/discussion/5881",
  "author_name": "",
  "post_date": "2013-09-26T00:45:25.497Z",
  "votes": 3,
  "comment_count": 2,
  "views": 3588,
  "content": "<p>Please let us know if you find <a href=\"https://www.kaggle.com/wiki/Leakage\">leakage</a>&nbsp;in the test set. We'll give you special profile badge as a token of thanks. Since this competition is meant to last for a few months, we'll all have more fun if there are no unfair hints that take the science out of it. Thanks!</p>",
  "messages": [
    {
      "id": "31574",
      "postDate": "09/26/2013 00:45:25",
      "content": "<p>Please let us know if you find <a href=\"https://www.kaggle.com/wiki/Leakage\">leakage</a>&nbsp;in the test set. We'll give you special profile badge as a token of thanks. Since this competition is meant to last for a few months, we'll all have more fun if there are no unfair hints that take the science out of it. Thanks!</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "31810",
      "postDate": "09/30/2013 01:48:22",
      "content": "<p>In my preliminary experimenting it seems images with cats are less complex than those with dogs.</p>\n<p>I'm just extracting simple point features (surf) at present without building up any dictionaries or anything, but it seems cats appear in images with (in my s/w -- a much-hacked version of OpenCV's &quot;find object&quot; sample program) an avg 60 significant pointlike features but dogs in images with 70.</p>\n<p>I suspect dogs tend to be photographed outside and cats are photographed in closeup. There may be 15-20 percent points in that.</p>\n<p>There is also the prospect of a few percent points in &nbsp;the image size. I suspect because of the &nbsp;the considerations above, larger images feature more cats than dogs. Even a pixel or 2 diff might leave a trace of which shelter or pet shop web page images originated from.</p>\n<p>It may be worthwhile making all the images the same size with some equalisation or other twiddling (e.g. moderate blur and/or sharpening) to reduce any inside/outside or near/far artifacts.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "31830",
      "postDate": "09/30/2013 15:16:30",
      "content": "<p>Thanks for the reply. We did already slightly modify the image dimensions to prevent a situation like &quot;Even a pixel or 2 diff might leave a trace of which shelter or pet shop web page images originated from.&quot;</p>\n<p>I'm not sure we can account for the way the photos are framed, and similarly we don't want to resize them and introduce artifacts. &nbsp;The complexity of the images is more an interesting observation than it is a source of leakage. &nbsp;Seems like it's so-far-so-good here.</p>",
      "rawMarkdown": "",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 31810,
      "author_name": "kymhorsell1",
      "author_url": "",
      "post_date": "09/30/2013 01:48:22",
      "content": "<p>In my preliminary experimenting it seems images with cats are less complex than those with dogs.</p>\n<p>I'm just extracting simple point features (surf) at present without building up any dictionaries or anything, but it seems cats appear in images with (in my s/w -- a much-hacked version of OpenCV's &quot;find object&quot; sample program) an avg 60 significant pointlike features but dogs in images with 70.</p>\n<p>I suspect dogs tend to be photographed outside and cats are photographed in closeup. There may be 15-20 percent points in that.</p>\n<p>There is also the prospect of a few percent points in &nbsp;the image size. I suspect because of the &nbsp;the considerations above, larger images feature more cats than dogs. Even a pixel or 2 diff might leave a trace of which shelter or pet shop web page images originated from.</p>\n<p>It may be worthwhile making all the images the same size with some equalisation or other twiddling (e.g. moderate blur and/or sharpening) to reduce any inside/outside or near/far artifacts.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 31830,
      "author_name": "wcukierski",
      "author_url": "",
      "post_date": "09/30/2013 15:16:30",
      "content": "<p>Thanks for the reply. We did already slightly modify the image dimensions to prevent a situation like &quot;Even a pixel or 2 diff might leave a trace of which shelter or pet shop web page images originated from.&quot;</p>\n<p>I'm not sure we can account for the way the photos are framed, and similarly we don't want to resize them and introduce artifacts. &nbsp;The complexity of the images is more an interesting observation than it is a source of leakage. &nbsp;Seems like it's so-far-so-good here.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "31574": "",
    "31810": "",
    "31830": ""
  },
  "source": "meta"
}