Grasp Detection on Cornell Grasping Dataset (Object-wise split)
88.7Point Grasp Success RateSAE, L1 reg.
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| SAE, L1 reg.Regularization=L12013.01 | 88.7 | 71.4 | — | |
| SAE, struct. reg. two-stageRegularization=Structured (Multimodal Group), Architecture=Two-stage cascaded2013.01 | 88.1 | 75.6 | — | |
| SAE, struct. reg., 2nd pass onlyRegularization=Structured (Multimodal Group), Pass=2nd only2013.01 | 87.6 | 73.2 | — | |
| SAE, struct. reg., 1st pass onlyRegularization=Structured (Multimodal Group), Pass=1st only2013.01 | 85.2 | 64.9 | — | |
| Jiang et al.2013.01 | 74.9 | 58.3 | — | |
| SAE, separate layer-1 feat.Features=Separate layer-12013.01 | 70.7 | 40 | — | |
| SAE, no mask-based scalingScaling=No mask-based2013.01 | 56.2 | 35.4 | — | |
| Chance2013.01 | 35.9 | 6.7 | — | |
| Chance2014.12 | — | — | 6.7 | |
| Direct Regression2014.12 | — | — | 84.9 | |
| Jiang et al.2014.12 | — | — | 58.3 | |
| Lenz et al.Time / image=13.5 sec2014.12 | — | — | 75.6 | |
| MultiGrasp Detection2014.12 | — | — | 87.1 | |
| Regression + ClassificationTime / image=76 ms2014.12 | — | — | 84.9 |