{
  "id": 180707,
  "title": "Understanding image output of AgentDataset ",
  "url": "/competitions/lyft-motion-prediction-autonomous-vehicles/discussion/180707",
  "author_name": "",
  "post_date": "2020-09-06T07:26:39.972646800Z",
  "votes": 1,
  "comment_count": 1,
  "views": 0,
  "content": "<p>What data is present in each of the 5 channels of the output image in the AgentDataset? </p>\n<p>When frame_history is set to 0, the output of an image in a the AgentDataset is an array of shape (5x224x224). </p>\n<p>Upon visualization, I see the first two channels contain masks and the last three channels are an RGB image of road layout information (road, lanes, and crosswalk). What does each masked channel indicate? </p>",
  "messages": [
    {
      "id": "999987",
      "postDate": "09/06/2020 07:26:39",
      "content": "<p>What data is present in each of the 5 channels of the output image in the AgentDataset? </p>\n<p>When frame_history is set to 0, the output of an image in a the AgentDataset is an array of shape (5x224x224). </p>\n<p>Upon visualization, I see the first two channels contain masks and the last three channels are an RGB image of road layout information (road, lanes, and crosswalk). What does each masked channel indicate? </p>",
      "rawMarkdown": "What data is present in each of the 5 channels of the output image in the AgentDataset? \n\nWhen frame_history is set to 0, the output of an image in a the AgentDataset is an array of shape (5x224x224). \n\nUpon visualization, I see the first two channels contain masks and the last three channels are an RGB image of road layout information (road, lanes, and crosswalk). What does each masked channel indicate?",
      "votes": null
    },
    {
      "id": "1000064",
      "postDate": "09/06/2020 08:50:30",
      "content": "<p>agents and ego vehicle. <br>\nif you increase the history, the layers for other agents and for the ego vehicle will increase accordingly.</p>\n<p>discussed <a href=\"https://www.kaggle.com/c/lyft-motion-prediction-autonomous-vehicles/discussion/178097\" target=\"_blank\">here</a></p>",
      "rawMarkdown": "agents and ego vehicle. \nif you increase the history, the layers for other agents and for the ego vehicle will increase accordingly.\n\ndiscussed [here](https://www.kaggle.com/c/lyft-motion-prediction-autonomous-vehicles/discussion/178097)",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1000064,
      "author_name": "ilu000",
      "author_url": "",
      "post_date": "09/06/2020 08:50:30",
      "content": "<p>agents and ego vehicle. <br>\nif you increase the history, the layers for other agents and for the ego vehicle will increase accordingly.</p>\n<p>discussed <a href=\"https://www.kaggle.com/c/lyft-motion-prediction-autonomous-vehicles/discussion/178097\" target=\"_blank\">here</a></p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "999987": "What data is present in each of the 5 channels of the output image in the AgentDataset? \n\nWhen frame_history is set to 0, the output of an image in a the AgentDataset is an array of shape (5x224x224). \n\nUpon visualization, I see the first two channels contain masks and the last three channels are an RGB image of road layout information (road, lanes, and crosswalk). What does each masked channel indicate?",
    "1000064": "agents and ego vehicle. \nif you increase the history, the layers for other agents and for the ego vehicle will increase accordingly.\n\ndiscussed [here](https://www.kaggle.com/c/lyft-motion-prediction-autonomous-vehicles/discussion/178097)"
  },
  "source": "meta"
}