Pose Estimation On Coco

評価結果

このベンチマークにおける各モデルのパフォーマンス結果

モデル名	AP	Paper Title
PATH (Partial FT)	77.1	HumanBench: Towards General Human-centric Perception with Projector Assisted Pretraining
BUCTD (PETR, with generative sampling)	-	Rethinking pose estimation in crowds: overcoming the detection information-bottleneck and ambiguity
PEFORMER-Xcit-dino-p8	72.6	PE-former: Pose Estimation Transformer
SOLIDER (swin-B)	76.6	Beyond Appearance: a Semantic Controllable Self-Supervised Learning Framework for Human-Centric Visual Tasks
MetaPrompt-SD	79.0	Harnessing Diffusion Models for Visual Perception with Meta Prompts
BUCTD (PETR, with generative sampling)	77.8	Rethinking pose estimation in crowds: overcoming the detection information-bottleneck and ambiguity
Hulk(Finetune, ViT-L)	78.7	Hulk: A Universal Knowledge Translator for Human-Centric Tasks
Hulk(Finetune, ViT-B)	77.5	Hulk: A Universal Knowledge Translator for Human-Centric Tasks
OmniPose (WASPv2)	79.5	OmniPose: A Multi-Scale Framework for Multi-Person Pose Estimation
I²R-Net (1st stage:HRFormer-B)	77.3	I^2R-Net: Intra- and Inter-Human Relation Network for Multi-Person Pose Estimation

0 of 10 row(s) selected.