PolarNet: 3D Point Clouds for Language-Guided Robotic Manipulation
Open paper
Benchmark
Model
Rank
Results
robot-manipulation-generalization-on-gembench
PolarNet
#6
Average Success Rate: 38.4
Average Success Rate (L1): 77.7±0.9
…
robot-manipulation-on-rlbench
PolarNet
#12
Succ. Rate (18 tasks, 100 demo/task): 46.4
…
Rank counts only results with a code link.