Joint-level Kinematic Estimation on PartNet-Mobility 3-part objects (test) benchmark leaderboard · SOTA2 Research

Joint-level Kinematic Estimation on PartNet-Mobility 3-part objects (test)

0Axis Angle Error 0PD2GS

Evaluation Results

Method	Links
PD2GS2025.06		0	0.34	0.26	—	—	0.03	0	0	0.58
PD2GS2025.06		0.01	0.12	0.29	—	0.03	0.12	0	0.12	0.45
ArtGS2025.06		0.02	0.85	0.03	—	—	0.14	0.03	0.01	0.02
ArtGS2025.06		0.02	89.03	14.64	—	0.57	1.79	0.02	75.13	13.43
PD2GS2025.06		0.23	1.34	0.19	0.03	0.23	0.14	1.01	1.59	2.83
ArtGS2025.06		0.29	0.05	0.21	0.19	0.01	0.39	15.89	0.17	6.02
DTArt2025.06		0.38	0.22	0.19	—	0.21	0.01	0.04	0.14	0.48
ArticulatedGS2025.06		0.52	6.53	3.62	0.83	8.32	0.42	5.23	2.13	14.43
ArticulatedGS2025.06		0.62	5.31	6.24	—	5.31	0.52	0.54	4.24	13.12
PARIS2025.06		0.77	87.4	76.58	—	—	0.22	0.15	0.17	87.02
ArticulatedGS2025.06		1.34	4.24	2.13	—	—	0.73	0.42	0.14	0.84
PARIS2025.06		1.69	89.4	79.91	—	0.52	0.84	0.18	157.14	109.55
DTArt2025.06		4.23	80.96	74.26	—	—	12.52	0.01	0.19	50.65
PD2GS2025.06		9.82	1.56	0.56	0.27	0.03	—	9.79	1.45	0.07
PARIS2025.06		9.96	29.61	0.47	0.13	0.31	0.3	30.23	111.21	33.72
ArticulatedGS2025.06		20.43	4.12	2.32	1.31	0.21	—	13.11	3.12	1.31
DTArt2025.06		27.6	0.2	78.01	4,601.75	0.17	—	59.95	0.16	0.37
ArtGS2025.06		54.17	1.68	0.75	0.29	0.08	—	96.54	1.66	0.02
PARIS2025.06		64.47	59.08	38.69	0.29	0.13	—	116.73	109.94	0.32
DTArt2025.06		88.32	62.54	64.68	1.99	5.56	6.79	99.97	67.17	82