{
  "id": 232698,
  "title": "GPS coordinates (obfuscated within 1km) released for WCS camera locations!",
  "url": "/competitions/iwildcam2021-fgvc8/discussion/232698",
  "author_name": "sbeery",
  "post_date": "2021-04-15T00:33:42.643000",
  "votes": 2,
  "comment_count": 0,
  "views": 0,
  "content": "<p>In order to allow competitors to try to use the geographic location of the cameras to improve their classification (<a href=\"https://arxiv.org/abs/1906.05272\" target=\"_blank\">as seen in this paper</a>), we have worked with WCS to release obfuscated GPS coordinates for most of the camera trap locations!  These can be downloaded on the competition <a href=\"https://github.com/visipedia/iwildcam_comp\" target=\"_blank\">github page</a>.  The precise coordinates of the cameras have been obfuscated randomly to within 1 km for privacy and security reasons, and correspond to the centers of the provided remote sensing imagery. Some of the obfuscated GPS locations (all from one country) were not released at the request of WCS, but knowing that the locations not listed in the <code>gps_locations.json</code> file are all from the same country should help competitors narrow down the set of possible species for those locations based on what is seen in the training data.</p>\n<p>This type of geographic reasoning about the set of species one might see in a camera is completely viable in a real-world scenarios, so we wanted to add this feature to make the competition even more realistic!</p>",
  "messages": [
    {
      "id": 1274070,
      "postDate": "2021-04-15T00:33:42.643Z",
      "content": "<p>In order to allow competitors to try to use the geographic location of the cameras to improve their classification (<a href=\"https://arxiv.org/abs/1906.05272\" target=\"_blank\">as seen in this paper</a>), we have worked with WCS to release obfuscated GPS coordinates for most of the camera trap locations!  These can be downloaded on the competition <a href=\"https://github.com/visipedia/iwildcam_comp\" target=\"_blank\">github page</a>.  The precise coordinates of the cameras have been obfuscated randomly to within 1 km for privacy and security reasons, and correspond to the centers of the provided remote sensing imagery. Some of the obfuscated GPS locations (all from one country) were not released at the request of WCS, but knowing that the locations not listed in the <code>gps_locations.json</code> file are all from the same country should help competitors narrow down the set of possible species for those locations based on what is seen in the training data.</p>\n<p>This type of geographic reasoning about the set of species one might see in a camera is completely viable in a real-world scenarios, so we wanted to add this feature to make the competition even more realistic!</p>",
      "rawMarkdown": "In order to allow competitors to try to use the geographic location of the cameras to improve their classification ([as seen in this paper](https://arxiv.org/abs/1906.05272)), we have worked with WCS to release obfuscated GPS coordinates for most of the camera trap locations!  These can be downloaded on the competition [github page](https://github.com/visipedia/iwildcam_comp).  The precise coordinates of the cameras have been obfuscated randomly to within 1 km for privacy and security reasons, and correspond to the centers of the provided remote sensing imagery. Some of the obfuscated GPS locations (all from one country) were not released at the request of WCS, but knowing that the locations not listed in the `gps_locations.json` file are all from the same country should help competitors narrow down the set of possible species for those locations based on what is seen in the training data.\n\nThis type of geographic reasoning about the set of species one might see in a camera is completely viable in a real-world scenarios, so we wanted to add this feature to make the competition even more realistic!",
      "votes": 2
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "1274070": "In order to allow competitors to try to use the geographic location of the cameras to improve their classification ([as seen in this paper](https://arxiv.org/abs/1906.05272)), we have worked with WCS to release obfuscated GPS coordinates for most of the camera trap locations!  These can be downloaded on the competition [github page](https://github.com/visipedia/iwildcam_comp).  The precise coordinates of the cameras have been obfuscated randomly to within 1 km for privacy and security reasons, and correspond to the centers of the provided remote sensing imagery. Some of the obfuscated GPS locations (all from one country) were not released at the request of WCS, but knowing that the locations not listed in the `gps_locations.json` file are all from the same country should help competitors narrow down the set of possible species for those locations based on what is seen in the training data.\n\nThis type of geographic reasoning about the set of species one might see in a camera is completely viable in a real-world scenarios, so we wanted to add this feature to make the competition even more realistic!"
  }
}