{
  "id": 500366,
  "title": "XFeat: Accelerated Features for Lightweight Image Matching",
  "url": "/competitions/image-matching-challenge-2024/discussion/500366",
  "author_name": "",
  "post_date": "2024-05-05T09:14:39.550796700Z",
  "votes": 25,
  "comment_count": 2,
  "views": 0,
  "content": "<p>Here is a new repository released called <strong>XFeat</strong>, claiming to be a fast and efficient keypoint detector and descriptor designed for real-time image matching tasks. They claim outperforming results when compared to the traditional methods like SIFT. There is a pretrained model and the repository is under Apache2.0 license. I wanted to share it with you as I thought it might be relevant to the context of the competition. Taking a look could be beneficial.</p>\n<p><img src=\"https://raw.githubusercontent.com/verlab/accelerated_features/main/figs/xfeat_quali.jpg\"></p>\n<p><strong>Paper:</strong> <a href=\"https://arxiv.org/abs/2404.19174\" target=\"_blank\">https://arxiv.org/abs/2404.19174</a><br>\n<strong>GitHub:</strong> <a href=\"https://github.com/verlab/accelerated_features\" target=\"_blank\">https://github.com/verlab/accelerated_features</a></p>",
  "messages": [
    {
      "id": "2794408",
      "postDate": "05/05/2024 09:14:39",
      "content": "<p>Here is a new repository released called <strong>XFeat</strong>, claiming to be a fast and efficient keypoint detector and descriptor designed for real-time image matching tasks. They claim outperforming results when compared to the traditional methods like SIFT. There is a pretrained model and the repository is under Apache2.0 license. I wanted to share it with you as I thought it might be relevant to the context of the competition. Taking a look could be beneficial.</p>\n<p><img src=\"https://raw.githubusercontent.com/verlab/accelerated_features/main/figs/xfeat_quali.jpg\"></p>\n<p><strong>Paper:</strong> <a href=\"https://arxiv.org/abs/2404.19174\" target=\"_blank\">https://arxiv.org/abs/2404.19174</a><br>\n<strong>GitHub:</strong> <a href=\"https://github.com/verlab/accelerated_features\" target=\"_blank\">https://github.com/verlab/accelerated_features</a></p>",
      "rawMarkdown": "Here is a new repository released called **XFeat**, claiming to be a fast and efficient keypoint detector and descriptor designed for real-time image matching tasks. They claim outperforming results when compared to the traditional methods like SIFT. There is a pretrained model and the repository is under Apache2.0 license. I wanted to share it with you as I thought it might be relevant to the context of the competition. Taking a look could be beneficial.\n\n![](https://raw.githubusercontent.com/verlab/accelerated_features/main/figs/xfeat_quali.jpg)\n\n**Paper:** https://arxiv.org/abs/2404.19174\n**GitHub:** https://github.com/verlab/accelerated_features",
      "votes": null
    },
    {
      "id": "2797567",
      "postDate": "05/06/2024 19:16:14",
      "content": "<p>Thank you for sharing this. :)</p>",
      "rawMarkdown": "Thank you for sharing this. :)",
      "votes": null
    },
    {
      "id": "2804778",
      "postDate": "05/10/2024 07:52:06",
      "content": "<p>Thank you for sharing.😀</p>",
      "rawMarkdown": "Thank you for sharing.😀",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2797567,
      "author_name": "junseonglee11",
      "author_url": "",
      "post_date": "05/06/2024 19:16:14",
      "content": "<p>Thank you for sharing this. :)</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 2804778,
      "author_name": "chenzhuoczczcz",
      "author_url": "",
      "post_date": "05/10/2024 07:52:06",
      "content": "<p>Thank you for sharing.😀</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2794408": "Here is a new repository released called **XFeat**, claiming to be a fast and efficient keypoint detector and descriptor designed for real-time image matching tasks. They claim outperforming results when compared to the traditional methods like SIFT. There is a pretrained model and the repository is under Apache2.0 license. I wanted to share it with you as I thought it might be relevant to the context of the competition. Taking a look could be beneficial.\n\n![](https://raw.githubusercontent.com/verlab/accelerated_features/main/figs/xfeat_quali.jpg)\n\n**Paper:** https://arxiv.org/abs/2404.19174\n**GitHub:** https://github.com/verlab/accelerated_features",
    "2797567": "Thank you for sharing this. :)",
    "2804778": "Thank you for sharing.😀"
  },
  "source": "meta"
}