Pose Estimation

Best model: Nate

Best model: HRNet fine-tuned on BRACE

Best model: I²R-Net (1st stage:HRFormer-B)

Best model: LOGO-CAP (Ours) HRNet-W48

Best model: MSPN

Best model: ViTPose (ViTAE-G, ensemble)

Best model: MogaNet-B (384x288)

Best model: BUCTD-W48 (w/cond. input from PETR, and generative sampling)

Best model: Parsing R-CNN + ResNext101

Best model: Stacked Hourglass Networks

Best model: Stacked Hourglass Networks

Best model: GIM-DKM

Best model: AdaPose

Best model: DECA-D3

Best model: SimpleBaseline + HANet

Best model: GeoNet

Best model: OmniPose

Best model: SPIGA

Best model: OmniPose (WASPv2)

Best model: PCT (swin-l, test set)

Best model: 4xRSN-50

Best model: UniHCP (finetune)

Best model: HQNet (ViT-L)

Best model: Mid-Level based

Best model: SubdivNet

Best model: AlphaPose

Best model: OmniPose