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

3D Human Pose Estimation on CMU Panoptic (test)

10.6MPJPE

Triangulation with Transformer Regressor (TTR) + Gröbner basis Corrector (GC) + Temporal Equivariant Rectifier (TER)

Updated 26d ago

Evaluation Results

Method
Triangulation with Transformer Regressor (TTR) + Gröbner basis Corrector (GC) + Temporal Equivariant Rectifier (TER) 2026.04	10.6	-	-	-	-	-	-	-	-	-	-	-	-
Zhang et al. 2026.04	11.2	-	-	-	-	-	-	-	-	-	-	-	-
Triangulation with Transformer Regressor (TTR) + Gröbner basis Corrector (GC) 2026.04	12.8	-	-	-	-	-	-	-	-	-	-	-	-
Iskakov et al. Volumetric 2026.04	13.7	-	-	-	-	-	-	-	-	-	-	-	-
LiCamPose 2023.12	14.44	-	-	-	-	-	-	-	-	-	-	-	11.61
VoxelPose(PRN) 2023.12	14.91	-	-	-	-	-	-	-	-	-	-	-	11.88
Chharia et al. 2026.04	15.3	-	-	-	-	-	-	-	-	-	-	-	-
Liao et al. 2026.04	15.6	-	-	-	-	-	-	-	-	-	-	-	-
Ye et al. 2026.04	17.9	-	-	-	-	-	-	-	-	-	-	-	-
PlanePose 2023.12	18.54	-	-	-	-	-	-	-	-	-	-	-	11.92
Iskakov et al. Algebraic 2026.04	21.3	-	-	-	-	-	-	-	-	-	-	-	-
LiCamPose 2023.12	22	-	-	-	-	-	-	-	-	-	-	-	15.96
Jiang et al. 2026.04	24.2	-	-	-	-	-	-	-	-	-	-	-	-
Bartol et al. 2026.04	25.4	-	-	-	-	-	-	-	-	-	-	-	-
MvP 2023.12	25.78	-	-	-	-	-	-	-	-	-	-	-	25.02
Gordon et al. 2026.04	28.9	-	-	-	-	-	-	-	-	-	-	-	-
DLT 2023.12	38.26	-	-	-	-	-	-	-	-	-	-	-	44.77
Distribution-Aware Single-stage 2022.03	53.8	-	53.3	51.2	49.1	61.5	-	-	-	-	-	-	-
DAS 2022.03	54.6	75	-	-	-	-	-	-	-	-	-	-	-
Ours (w/ Synth) 2022.07	55	-	55.2	55	50.4	61.4	-	-	-	-	-	-	-
Wang et al. 2022.03	55.1	-	50.9	50.5	50.7	68.2	-	-	-	-	-	-	-
Ours 2022.07	56.1	-	54.7	55.2	50.1	66.4	-	-	-	-	-	-	-
Zhen et al. 2022.03	61.8	108	-	-	-	-	-	-	-	-	-	-	-
Zhen et al. 2022.03	61.8	-	63.1	60.3	56.6	67.1	-	-	-	-	-	-	-
SMAP 2022.07	61.8	-	63.1	60.3	56.6	67.1	-	-	-	-	-	-	-
Fabbri et al. 2022.03	69	125	-	-	-	-	-	-	-	-	-	-	-
Fabbri et al. 2022.03	69	-	45	95	58	79	-	-	-	-	-	-	-
Zanfir et al. 2022.07	72.1	-	72.4	78.8	66.8	94.3	-	-	-	-	-	-	-
Zanfir et al. [45] 2022.03	78.1	-	72.4	78.8	66.8	94.3	-	-	-	-	-	-	-
Moon et al. 2022.07	87.6	-	89.6	91.3	79.6	90.1	-	-	-	-	-	-	-
Zanfir et al. [44] 2022.03	153.4	-	140	165.9	150.7	156	-	-	-	-	-	-	-
Popa et al. 2022.03	203.4	-	217.9	187.3	193.6	221.3	-	-	-	-	-	-	-
Moon et al. 2022.03	-	107	-	-	-	-	-	-	-	-	-	-	-
Lin et al. 2022.03	-	118	-	-	-	-	-	-	-	-	-	-	-
Learnable Triangulation 2024.04	-	-	-	-	-	-	80.6	716.1	-	-	-	-	-
Learnable Triangulation 2024.04	-	-	-	-	-	-	80.6	1,326.9	-	-	-	-	-
Epipolar Transformers 2024.04	-	-	-	-	-	-	68.1	406.5	-	-	-	-	-
Epipolar Transformers 2024.04	-	-	-	-	-	-	68.1	1,016.2	-	-	-	-	-
MTF-Transformers 2024.04	-	-	-	-	-	-	78.6	407	-	-	-	-	-
MTF-Transformers 2024.04	-	-	-	-	-	-	78.6	1,017.8	-	-	-	-	-
AdaFuse 2024.04	-	-	-	-	-	-	69.7	595	-	-	-	-	-
AdaFuse 2024.04	-	-	-	-	-	-	69.7	1,487.6	-	-	-	-	-
UPose3D 2024.04	-	-	-	-	-	-	65.4	208.7	-	-	-	-	-
UPose3D 2024.04	-	-	-	-	-	-	65.4	517.7	-	-	-	-	-
2D HPE + ml-depth-pro 2025.12	-	-	-	-	-	-	-	-	215.9	-	-	-	-
2D HPE + DepthAnything2 2025.12	-	-	-	-	-	-	-	-	248.6	-	-	-	-
RUMPL 2025.12	-	-	-	-	-	-	-	-	95.9	-	-	-	-
Learn.-triang. 2025.12	-	-	-	-	-	-	-	-	130.9	-	-	-	-
PPT 2025.12	-	-	-	-	-	-	-	-	108.3	114.2	-	-	-
Triangulation 2025.12	-	-	-	-	-	-	-	-	44	43	-	-	-
FullyConnected + RCS + Rays 2025.12	-	-	-	-	-	-	-	-	127.4	108.6	-	-	-
MPL + RCS 2025.12	-	-	-	-	-	-	-	-	278.8	274.9	-	-	-
MPL + RCS + Rays 2025.12	-	-	-	-	-	-	-	-	55.7	45.4	-	-	-
RUMPL (w/o Conf) 2025.12	-	-	-	-	-	-	-	-	41.1	36.6	-	-	-
RUMPL 2025.12	-	-	-	-	-	-	-	-	35	30.8	-	-	-
WHENet 2024.06	-	-	-	-	-	-	-	-	-	-	29.55	-	-
HopeNet 2024.06	-	-	-	-	-	-	-	-	-	-	22.16	-	-
6DRepNet 2024.06	-	-	-	-	-	-	-	-	-	-	10.74	-	-
PoseBERT 2024.06	-	-	-	-	-	-	-	-	-	-	22.32	-	-
Frozen CogVLM + Regressor 2024.06	-	-	-	-	-	-	-	-	-	-	32.11	-	-
Frozen CogVLM + Regressor 2024.06	-	-	-	-	-	-	-	-	-	-	33.47	-	-
Vision Tower (ViT) + Regressor 2024.06	-	-	-	-	-	-	-	-	-	-	10.88	-	-
Non-merging CogVLM 2024.06	-	-	-	-	-	-	-	-	-	-	8.18	0.13	-
TA merging CogVLM 2024.06	-	-	-	-	-	-	-	-	-	-	7.72	68.9	-
HPE-CogVLM 2024.06	-	-	-	-	-	-	-	-	-	-	7.36	0.052	-