Update README.md
Browse files
README.md
CHANGED
|
@@ -33,26 +33,20 @@ Below one can find the description of each field in the dataset.
|
|
| 33 |
|
| 34 |
- **`question_id` (str)**
|
| 35 |
Unique identifier for each question–agent pair.
|
| 36 |
-
|
| 37 |
- **`question` (str)**
|
| 38 |
Natural language navigation question describing whether an agent can move from a start area to a goal area.
|
| 39 |
-
|
| 40 |
- **`scene_id` (str)**
|
| 41 |
Identifier of the indoor scene (e.g., `HM3D00000`, `MP3D00027`).
|
| 42 |
-
|
| 43 |
- **`scene_type` (str)**
|
| 44 |
High-level category of the scene (e.g., `home`).
|
| 45 |
-
|
| 46 |
- **`scene_nodes` (list[dict])**
|
| 47 |
List of scene graph nodes available in the environment.
|
| 48 |
Each node contains:
|
| 49 |
- `node_id` (str): Unique identifier of the node
|
| 50 |
- `name` (str): Human-readable node name (e.g., room or area label)
|
| 51 |
-
|
| 52 |
- **`agent_name` (str)**
|
| 53 |
Name of the agent under which the navigation question should be evaluated.
|
| 54 |
This field links to an entry in the Agent Profiles Dataset.
|
| 55 |
-
|
| 56 |
- **`answer` (optional / null)**
|
| 57 |
Ground-truth navigation feasibility label.
|
| 58 |
This field is intentionally left empty in the current release and may be populated in future versions.
|
|
@@ -69,38 +63,27 @@ This dataset defines the physical properties and functional capabilities of each
|
|
| 69 |
used in the CapNav benchmark.
|
| 70 |
|
| 71 |
#### Data Fields
|
| 72 |
-
|
| 73 |
- **`agent_name` (str)**
|
| 74 |
Unique identifier of the agent (e.g., `HUMAN`, `WHEELCHAIR`, `SWEEPER`).
|
| 75 |
-
|
| 76 |
- **`body_shape` (str)**
|
| 77 |
Abstract geometric representation of the agent’s body (e.g., `cylinder`, `box`).
|
| 78 |
-
|
| 79 |
- **`body_height_m` (float)**
|
| 80 |
Agent height in meters.
|
| 81 |
-
|
| 82 |
- **`body_width_m` (float)**
|
| 83 |
Agent width in meters.
|
| 84 |
-
|
| 85 |
- **`body_depth_m` (float, optional)**
|
| 86 |
Agent depth in meters.
|
| 87 |
May be `null` for rotationally symmetric agents.
|
| 88 |
-
|
| 89 |
- **`max_vertical_cross_height_m` (float)**
|
| 90 |
Maximum vertical obstacle height the agent can cross.
|
| 91 |
-
|
| 92 |
- **`can_go_up_or_down_stairs` (bool)**
|
| 93 |
Indicates whether the agent is capable of traversing stairs.
|
| 94 |
-
|
| 95 |
- **`can_operate_elevator` (bool)**
|
| 96 |
Indicates whether the agent can operate and use elevators.
|
| 97 |
-
|
| 98 |
- **`can_open_the_door` (bool)**
|
| 99 |
Indicates whether the agent can open doors independently.
|
| 100 |
-
|
| 101 |
- **`description` (str)**
|
| 102 |
Natural language description summarizing the agent’s physical and functional characteristics.
|
| 103 |
-
|
| 104 |
---
|
| 105 |
|
| 106 |
## Relationship Between the Datasets
|
|
|
|
| 33 |
|
| 34 |
- **`question_id` (str)**
|
| 35 |
Unique identifier for each question–agent pair.
|
|
|
|
| 36 |
- **`question` (str)**
|
| 37 |
Natural language navigation question describing whether an agent can move from a start area to a goal area.
|
|
|
|
| 38 |
- **`scene_id` (str)**
|
| 39 |
Identifier of the indoor scene (e.g., `HM3D00000`, `MP3D00027`).
|
|
|
|
| 40 |
- **`scene_type` (str)**
|
| 41 |
High-level category of the scene (e.g., `home`).
|
|
|
|
| 42 |
- **`scene_nodes` (list[dict])**
|
| 43 |
List of scene graph nodes available in the environment.
|
| 44 |
Each node contains:
|
| 45 |
- `node_id` (str): Unique identifier of the node
|
| 46 |
- `name` (str): Human-readable node name (e.g., room or area label)
|
|
|
|
| 47 |
- **`agent_name` (str)**
|
| 48 |
Name of the agent under which the navigation question should be evaluated.
|
| 49 |
This field links to an entry in the Agent Profiles Dataset.
|
|
|
|
| 50 |
- **`answer` (optional / null)**
|
| 51 |
Ground-truth navigation feasibility label.
|
| 52 |
This field is intentionally left empty in the current release and may be populated in future versions.
|
|
|
|
| 63 |
used in the CapNav benchmark.
|
| 64 |
|
| 65 |
#### Data Fields
|
|
|
|
| 66 |
- **`agent_name` (str)**
|
| 67 |
Unique identifier of the agent (e.g., `HUMAN`, `WHEELCHAIR`, `SWEEPER`).
|
|
|
|
| 68 |
- **`body_shape` (str)**
|
| 69 |
Abstract geometric representation of the agent’s body (e.g., `cylinder`, `box`).
|
|
|
|
| 70 |
- **`body_height_m` (float)**
|
| 71 |
Agent height in meters.
|
|
|
|
| 72 |
- **`body_width_m` (float)**
|
| 73 |
Agent width in meters.
|
|
|
|
| 74 |
- **`body_depth_m` (float, optional)**
|
| 75 |
Agent depth in meters.
|
| 76 |
May be `null` for rotationally symmetric agents.
|
|
|
|
| 77 |
- **`max_vertical_cross_height_m` (float)**
|
| 78 |
Maximum vertical obstacle height the agent can cross.
|
|
|
|
| 79 |
- **`can_go_up_or_down_stairs` (bool)**
|
| 80 |
Indicates whether the agent is capable of traversing stairs.
|
|
|
|
| 81 |
- **`can_operate_elevator` (bool)**
|
| 82 |
Indicates whether the agent can operate and use elevators.
|
|
|
|
| 83 |
- **`can_open_the_door` (bool)**
|
| 84 |
Indicates whether the agent can open doors independently.
|
|
|
|
| 85 |
- **`description` (str)**
|
| 86 |
Natural language description summarizing the agent’s physical and functional characteristics.
|
|
|
|
| 87 |
---
|
| 88 |
|
| 89 |
## Relationship Between the Datasets
|