Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

SpatialRGPT-Bench

Benchmarks

Task NameDataset NameSOTA ResultTrend
Spatial ReasoningSpatialRGPT-Bench qualitative 1.0 (val test)
Below/Above Accuracy99.17
11
Object-level 3D UnderstandingSpatialRGPT-Bench Quantitative
Direct Distance Accuracy35.1
4
Object-level 3D UnderstandingSpatialRGPT-Bench Qualitative
Below/Above Accuracy99.17
4
Showing 3 of 3 rows