3D-Aware VLMs with Implicit and Explicit Geometries
Open paper
Benchmark
Model
Rank
Results
3d-dense-captioning-on-scan2cap
VLM-IE3D
#2
CIDEr: 80.4
spatial-reasoning-on-vsi-bench
VLM-IE3D-4B (extra data)
#5
Accuracy: 47.6
Rank counts only results with a code link.