{
  "id": "2004.02857",
  "version": "v1",
  "published": "2020-04-06T17:49:12.000Z",
  "updated": "2020-04-06T17:49:12.000Z",
  "title": "Beyond the Nav-Graph: Vision-and-Language Navigation in Continuous Environments",
  "authors": [
    "Jacob Krantz",
    "Erik Wijmans",
    "Arjun Majumdar",
    "Dhruv Batra",
    "Stefan Lee"
  ],
  "categories": [
    "cs.CV",
    "cs.CL",
    "cs.RO"
  ],
  "abstract": "We develop a language-guided navigation task set in a continuous 3D environment where agents must execute low-level actions to follow natural language navigation directions. By being situated in continuous environments, this setting lifts a number of assumptions implicit in prior work that represents environments as a sparse graph of panoramas with edges corresponding to navigability. Specifically, our setting drops the presumptions of known environment topologies, short-range oracle navigation, and perfect agent localization. To contextualize this new task, we develop models that mirror many of the advances made in prior settings as well as single-modality baselines. While some of these techniques transfer, we find significantly lower absolute performance in the continuous setting -- suggesting that performance in prior `navigation-graph' settings may be inflated by the strong implicit assumptions.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2020-04-06T17:49:12.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "continuous environments",
      "vision-and-language navigation",
      "natural language navigation directions",
      "strong implicit assumptions",
      "language-guided navigation task set"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 0,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}