{
  "id": 23139,
  "title": "#70: Based on Skin Segmentation",
  "url": "/competitions/state-farm-distracted-driver-detection/writeups/heshameraqi-70-based-on-skin-segmentation",
  "author_name": "",
  "post_date": "2016-08-28T14:23:48.317Z",
  "votes": 8,
  "comment_count": 2,
  "views": 550,
  "content": "<p>Hello everyone, my system is not within the top 10, but 70th is not very bad. Especially that I classify each image independently (No semi-supervised learning).\nSo, I wanted to share my system details maybe they are useful to you :-)\nAny questions are welcomed.</p>\n\n<p><img src=\"https://s14.postimg.io/ama3i7q5t/image.png\" alt=\"enter image description here\" title>\n<strong>References:</strong>\n1 - Rajen Bhatt, Abhinav Dhall, 'Skin Segmentation Dataset', UCI Machine Learning Repository.\n2 - X. Zhu, D. Ramanan. &quot;Face detection, pose estimation and landmark localization in the wild&quot; Computer Vision and Pattern Recognition (CVPR) Providence, Rhode Island, June 2012.\n3 - Simonyan, Karen, and Andrew Zisserman. &quot;Very deep convolutional networks for large-scale image recognition.&quot; arXiv preprint arXiv:1409.1556 (2014).</p>\n\n<p><strong>Notes:</strong></p>\n\n<ul>\n<li>Face detection for such problem is hard, the well-known Haar Cascades surely fail.</li>\n<li>The example in the image above is easy, but normally driver face is seen from side.</li>\n<li>I've tried lots of strategies for upper body Human Pose Estimation (Calvin) and scene/driver segmentation (this and this) but didn't achieve good results.</li>\n<li>I used data augmentation. Because the training data size is not large.</li>\n</ul>",
  "messages": [
    {
      "id": "132691",
      "postDate": "08/28/2016 14:23:48",
      "content": "<p>Hello everyone, my system is not within the top 10, but 70th is not very bad. Especially that I classify each image independently (No semi-supervised learning).\nSo, I wanted to share my system details maybe they are useful to you :-)\nAny questions are welcomed.</p>\n\n<p><img src=\"https://s14.postimg.io/ama3i7q5t/image.png\" alt=\"enter image description here\" title>\n<strong>References:</strong>\n1 - Rajen Bhatt, Abhinav Dhall, 'Skin Segmentation Dataset', UCI Machine Learning Repository.\n2 - X. Zhu, D. Ramanan. &quot;Face detection, pose estimation and landmark localization in the wild&quot; Computer Vision and Pattern Recognition (CVPR) Providence, Rhode Island, June 2012.\n3 - Simonyan, Karen, and Andrew Zisserman. &quot;Very deep convolutional networks for large-scale image recognition.&quot; arXiv preprint arXiv:1409.1556 (2014).</p>\n\n<p><strong>Notes:</strong></p>\n\n<ul>\n<li>Face detection for such problem is hard, the well-known Haar Cascades surely fail.</li>\n<li>The example in the image above is easy, but normally driver face is seen from side.</li>\n<li>I've tried lots of strategies for upper body Human Pose Estimation (Calvin) and scene/driver segmentation (this and this) but didn't achieve good results.</li>\n<li>I used data augmentation. Because the training data size is not large.</li>\n</ul>",
      "rawMarkdown": "Hello everyone, my system is not within the top 10, but 70th is not very bad. Especially that I classify each image independently (No semi-supervised learning).\r\nSo, I wanted to share my system details maybe they are useful to you :-)\r\nAny questions are welcomed.\r\n\r\n![enter image description here][1]\r\n**References:**\r\n1 - Rajen Bhatt, Abhinav Dhall, 'Skin Segmentation Dataset', UCI Machine Learning Repository.\r\n2 - X. Zhu, D. Ramanan. \"Face detection, pose estimation and landmark localization in the wild\" Computer Vision and Pattern Recognition (CVPR) Providence, Rhode Island, June 2012.\r\n3 - Simonyan, Karen, and Andrew Zisserman. \"Very deep convolutional networks for large-scale image recognition.\" arXiv preprint arXiv:1409.1556 (2014).\r\n\r\n**Notes:**\r\n\r\n - Face detection for such problem is hard, the well-known Haar Cascades surely fail.\r\n - The example in the image above is easy, but normally driver face is seen from side.\r\n - I've tried lots of strategies for upper body Human Pose Estimation (Calvin) and scene/driver segmentation (this and this) but didn't achieve good results.\r\n - I used data augmentation. Because the training data size is not large.\r\n\r\n  [1]: https://s14.postimg.io/ama3i7q5t/image.png",
      "votes": null
    },
    {
      "id": "868759",
      "postDate": "05/31/2020 13:16:41",
      "content": "<p>How did you obtain the skin images?</p>",
      "rawMarkdown": "How did you obtain the skin images?",
      "votes": null
    },
    {
      "id": "2108929",
      "postDate": "01/21/2023 00:35:09",
      "content": "<p>I trained a Bayesian classifier on the UCI Skin Segmentation Dataset: <a href=\"https://archive.ics.uci.edu/ml/datasets/Skin+Segmentation\" target=\"_blank\">https://archive.ics.uci.edu/ml/datasets/Skin+Segmentation</a> . I earlier wrote a related blog post about that (discussing some other topic but using the same dataset): <a href=\"https://heraqi.blogspot.com/2016/06/does-more-features-always-mean-higher.html\" target=\"_blank\">https://heraqi.blogspot.com/2016/06/does-more-features-always-mean-higher.html</a></p>",
      "rawMarkdown": "I trained a Bayesian classifier on the UCI Skin Segmentation Dataset: https://archive.ics.uci.edu/ml/datasets/Skin+Segmentation . I earlier wrote a related blog post about that (discussing some other topic but using the same dataset): https://heraqi.blogspot.com/2016/06/does-more-features-always-mean-higher.html",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 868759,
      "author_name": "trisha73",
      "author_url": "",
      "post_date": "05/31/2020 13:16:41",
      "content": "<p>How did you obtain the skin images?</p>",
      "votes": null,
      "replies": [
        {
          "id": 2108929,
          "author_name": "heshameraqi",
          "author_url": "",
          "post_date": "01/21/2023 00:35:09",
          "content": "<p>I trained a Bayesian classifier on the UCI Skin Segmentation Dataset: <a href=\"https://archive.ics.uci.edu/ml/datasets/Skin+Segmentation\" target=\"_blank\">https://archive.ics.uci.edu/ml/datasets/Skin+Segmentation</a> . I earlier wrote a related blog post about that (discussing some other topic but using the same dataset): <a href=\"https://heraqi.blogspot.com/2016/06/does-more-features-always-mean-higher.html\" target=\"_blank\">https://heraqi.blogspot.com/2016/06/does-more-features-always-mean-higher.html</a></p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "132691": "Hello everyone, my system is not within the top 10, but 70th is not very bad. Especially that I classify each image independently (No semi-supervised learning).\r\nSo, I wanted to share my system details maybe they are useful to you :-)\r\nAny questions are welcomed.\r\n\r\n![enter image description here][1]\r\n**References:**\r\n1 - Rajen Bhatt, Abhinav Dhall, 'Skin Segmentation Dataset', UCI Machine Learning Repository.\r\n2 - X. Zhu, D. Ramanan. \"Face detection, pose estimation and landmark localization in the wild\" Computer Vision and Pattern Recognition (CVPR) Providence, Rhode Island, June 2012.\r\n3 - Simonyan, Karen, and Andrew Zisserman. \"Very deep convolutional networks for large-scale image recognition.\" arXiv preprint arXiv:1409.1556 (2014).\r\n\r\n**Notes:**\r\n\r\n - Face detection for such problem is hard, the well-known Haar Cascades surely fail.\r\n - The example in the image above is easy, but normally driver face is seen from side.\r\n - I've tried lots of strategies for upper body Human Pose Estimation (Calvin) and scene/driver segmentation (this and this) but didn't achieve good results.\r\n - I used data augmentation. Because the training data size is not large.\r\n\r\n  [1]: https://s14.postimg.io/ama3i7q5t/image.png",
    "868759": "How did you obtain the skin images?",
    "2108929": "I trained a Bayesian classifier on the UCI Skin Segmentation Dataset: https://archive.ics.uci.edu/ml/datasets/Skin+Segmentation . I earlier wrote a related blog post about that (discussing some other topic but using the same dataset): https://heraqi.blogspot.com/2016/06/does-more-features-always-mean-higher.html"
  },
  "source": "meta"
}