{
  "id": 149211,
  "title": "Detecting with CenterNets",
  "url": "/competitions/global-wheat-detection/discussion/149211",
  "author_name": "",
  "post_date": "2020-05-07T10:31:33.696166900Z",
  "votes": 4,
  "comment_count": 1,
  "views": 0,
  "content": "<p><a href=\"https://www.kaggle.com/kyoshioka47/centernet-starterkit-pytorch\">https://www.kaggle.com/kyoshioka47/centernet-starterkit-pytorch</a>\nI made a starter kit notebook for training detectors based on CenterNet.</p>\n\n<h2>Why CenterNet?</h2>\n\n<p><img src=\"https://github.com/xingyizhou/CenterNet/raw/master/readme/fig2.png\" alt=\"\"></p>\n\n<p>1) CenterNet is simple and easy to customize compared to faster rcnns.</p>\n\n<p>2) CenterNet is fast, since it's a single-stage detector.</p>\n\n<p>3) Can be accurate as the faster rcnn if trained properly!</p>\n\n<p>Single-stage detectors are a lot simple and easy to customize compared to two-stage detectors.</p>\n\n<p>Wonder which one suits for the task?\nLet's find out on the way!</p>",
  "messages": [
    {
      "id": "836886",
      "postDate": "05/07/2020 10:31:33",
      "content": "<p><a href=\"https://www.kaggle.com/kyoshioka47/centernet-starterkit-pytorch\">https://www.kaggle.com/kyoshioka47/centernet-starterkit-pytorch</a>\nI made a starter kit notebook for training detectors based on CenterNet.</p>\n\n<h2>Why CenterNet?</h2>\n\n<p><img src=\"https://github.com/xingyizhou/CenterNet/raw/master/readme/fig2.png\" alt=\"\"></p>\n\n<p>1) CenterNet is simple and easy to customize compared to faster rcnns.</p>\n\n<p>2) CenterNet is fast, since it's a single-stage detector.</p>\n\n<p>3) Can be accurate as the faster rcnn if trained properly!</p>\n\n<p>Single-stage detectors are a lot simple and easy to customize compared to two-stage detectors.</p>\n\n<p>Wonder which one suits for the task?\nLet's find out on the way!</p>",
      "rawMarkdown": "https://www.kaggle.com/kyoshioka47/centernet-starterkit-pytorch\nI made a starter kit notebook for training detectors based on CenterNet.\n\n## Why CenterNet?\n\n![](https://github.com/xingyizhou/CenterNet/raw/master/readme/fig2.png)\n\n1) CenterNet is simple and easy to customize compared to faster rcnns.\n\n2) CenterNet is fast, since it's a single-stage detector.\n\n3) Can be accurate as the faster rcnn if trained properly!\n\nSingle-stage detectors are a lot simple and easy to customize compared to two-stage detectors.\n\nWonder which one suits for the task?\nLet's find out on the way!",
      "votes": null
    },
    {
      "id": "836910",
      "postDate": "05/07/2020 11:04:37",
      "content": "<p>Great one, looks amazing.\nThanks.</p>",
      "rawMarkdown": "Great one, looks amazing.\nThanks.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 836910,
      "author_name": "darshangandhi",
      "author_url": "",
      "post_date": "05/07/2020 11:04:37",
      "content": "<p>Great one, looks amazing.\nThanks.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "836886": "https://www.kaggle.com/kyoshioka47/centernet-starterkit-pytorch\nI made a starter kit notebook for training detectors based on CenterNet.\n\n## Why CenterNet?\n\n![](https://github.com/xingyizhou/CenterNet/raw/master/readme/fig2.png)\n\n1) CenterNet is simple and easy to customize compared to faster rcnns.\n\n2) CenterNet is fast, since it's a single-stage detector.\n\n3) Can be accurate as the faster rcnn if trained properly!\n\nSingle-stage detectors are a lot simple and easy to customize compared to two-stage detectors.\n\nWonder which one suits for the task?\nLet's find out on the way!",
    "836910": "Great one, looks amazing.\nThanks."
  },
  "source": "meta"
}