{
  "id": 224490,
  "title": "Welcome to Semi-iNat 2021 Challenge!",
  "url": "/competitions/semi-inat-2021/discussion/224490",
  "author_name": "Jong-Chyi Su",
  "post_date": "2021-03-08T17:45:30.403000",
  "votes": 5,
  "comment_count": 0,
  "views": 0,
  "content": "<p>Hi everyone,</p>\n<p>We are happy to announce the <strong>second semi-supervised fine-grained recognition challenge (Semi-iNat 2021)</strong>. The challenge is part of the Eighth Workshop on Fine-Grained Visual Categorization (<a href=\"https://sites.google.com/view/fgvc8\" target=\"_blank\">FGVC8</a>) at <a href=\"http://cvpr2021.thecvf.com/\" target=\"_blank\">CVPR 2021</a>.</p>\n<p>This year's challenge is different from the previous year (<a href=\"https://www.kaggle.com/c/semi-inat-2020\" target=\"_blank\">Semi-Aves challenge</a>) in the following aspects:</p>\n<ol>\n<li><strong>Combined unlabeled data:</strong> We combined unlabeled data from in-class and out-of-class, and do not provide the label.</li>\n<li><strong>New species:</strong> Our dataset has no overlapping species between previous iNat competitions, including iNat-17, iNat-18, iNat-19, iNat-21, and Semi-Aves.</li>\n<li><strong>Species from different kingdoms:</strong> Unlike Semi-Aves where all the species are from the Aves (birds) kingdom, we have a diverse set of species including animals, plants, and fungi. The distribution is shown in the following table.</li>\n</ol>\n<p>We are looking forward to your creative solutions and interesting findings during the competition. Feel free to ask questions here, or on the <a href=\"https://github.com/cvl-umass/semi-inat-2021\" target=\"_blank\">github</a> page. After the competition ends, we encourage you to share your solution and findings with the community. We will also invite top teams to present their work at the workshop.</p>",
  "messages": [
    {
      "id": 1231142,
      "postDate": "2021-03-08T17:45:30.403Z",
      "content": "<p>Hi everyone,</p>\n<p>We are happy to announce the <strong>second semi-supervised fine-grained recognition challenge (Semi-iNat 2021)</strong>. The challenge is part of the Eighth Workshop on Fine-Grained Visual Categorization (<a href=\"https://sites.google.com/view/fgvc8\" target=\"_blank\">FGVC8</a>) at <a href=\"http://cvpr2021.thecvf.com/\" target=\"_blank\">CVPR 2021</a>.</p>\n<p>This year's challenge is different from the previous year (<a href=\"https://www.kaggle.com/c/semi-inat-2020\" target=\"_blank\">Semi-Aves challenge</a>) in the following aspects:</p>\n<ol>\n<li><strong>Combined unlabeled data:</strong> We combined unlabeled data from in-class and out-of-class, and do not provide the label.</li>\n<li><strong>New species:</strong> Our dataset has no overlapping species between previous iNat competitions, including iNat-17, iNat-18, iNat-19, iNat-21, and Semi-Aves.</li>\n<li><strong>Species from different kingdoms:</strong> Unlike Semi-Aves where all the species are from the Aves (birds) kingdom, we have a diverse set of species including animals, plants, and fungi. The distribution is shown in the following table.</li>\n</ol>\n<p>We are looking forward to your creative solutions and interesting findings during the competition. Feel free to ask questions here, or on the <a href=\"https://github.com/cvl-umass/semi-inat-2021\" target=\"_blank\">github</a> page. After the competition ends, we encourage you to share your solution and findings with the community. We will also invite top teams to present their work at the workshop.</p>",
      "rawMarkdown": "Hi everyone,\n\nWe are happy to announce the **second semi-supervised fine-grained recognition challenge (Semi-iNat 2021)**. The challenge is part of the Eighth Workshop on Fine-Grained Visual Categorization ([FGVC8](https://sites.google.com/view/fgvc8)) at [CVPR 2021](http://cvpr2021.thecvf.com/).\n\nThis year's challenge is different from the previous year ([Semi-Aves challenge](https://www.kaggle.com/c/semi-inat-2020)) in the following aspects:\n1. **Combined unlabeled data:** We combined unlabeled data from in-class and out-of-class, and do not provide the label.\n2. **New species:** Our dataset has no overlapping species between previous iNat competitions, including iNat-17, iNat-18, iNat-19, iNat-21, and Semi-Aves.\n3. **Species from different kingdoms:** Unlike Semi-Aves where all the species are from the Aves (birds) kingdom, we have a diverse set of species including animals, plants, and fungi. The distribution is shown in the following table.\n\nWe are looking forward to your creative solutions and interesting findings during the competition. Feel free to ask questions here, or on the [github](https://github.com/cvl-umass/semi-inat-2021) page. After the competition ends, we encourage you to share your solution and findings with the community. We will also invite top teams to present their work at the workshop.",
      "votes": 5
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "1231142": "Hi everyone,\n\nWe are happy to announce the **second semi-supervised fine-grained recognition challenge (Semi-iNat 2021)**. The challenge is part of the Eighth Workshop on Fine-Grained Visual Categorization ([FGVC8](https://sites.google.com/view/fgvc8)) at [CVPR 2021](http://cvpr2021.thecvf.com/).\n\nThis year's challenge is different from the previous year ([Semi-Aves challenge](https://www.kaggle.com/c/semi-inat-2020)) in the following aspects:\n1. **Combined unlabeled data:** We combined unlabeled data from in-class and out-of-class, and do not provide the label.\n2. **New species:** Our dataset has no overlapping species between previous iNat competitions, including iNat-17, iNat-18, iNat-19, iNat-21, and Semi-Aves.\n3. **Species from different kingdoms:** Unlike Semi-Aves where all the species are from the Aves (birds) kingdom, we have a diverse set of species including animals, plants, and fungi. The distribution is shown in the following table.\n\nWe are looking forward to your creative solutions and interesting findings during the competition. Feel free to ask questions here, or on the [github](https://github.com/cvl-umass/semi-inat-2021) page. After the competition ends, we encourage you to share your solution and findings with the community. We will also invite top teams to present their work at the workshop."
  }
}