PIG-Nav: Key Insights for Pretrained Image Goal Navigation Models
Paper
•
2507.17220
•
Published
•
1
Search is not available for this dataset
image
imagewidth (px) 640
640
|
|---|
The 3D game environments for our paper PIG-Nav: Key Insights for Pretrained Image-Goal Navigation Models, used for model evaluation.
Description of the environments:
- The two environments, Highrise and Sanctuary, are developed from ShooterGame, which is a public project of Unreal Engine 4.27.
- We provide the game binaries in both Windows and Linux versions of Highrise and Sanctuary environments in this repo.
- We use UnrealCV for communications between game binaries and PIG-Nav agent. We implement a rule-based controller to transform delta actions into keypresses in games.
- Please follow our github repo https://github.com/zpschang/PIG-Nav for detailed use.
- Downloads last month
- 68
Size of downloaded dataset files:
1.84 MB
Size of the auto-converted Parquet files:
1.48 MB
Number of rows:
2