Command Palette
Search for a command to run...
Pose Estimation
Pose Estimation is a task in the field of computer vision that aims to detect the position and posture of people or objects. This task achieves human pose estimation by predicting the locations of specific keypoints (such as hands, head, elbows, etc.). Pose Estimation has significant application value in areas like human-computer interaction, motion analysis, and virtual reality. Common benchmark tests include the MPII Human Pose dataset.
MPII Human Pose
PCT (swin-l, test set)
COCO test-dev
HRNet-W48+DARK
Leeds Sports Poses
OmniPose
OCHuman
ViTPose (ViTAE-G, GT bounding boxes)
CrowdPose
BUCTD-W48 (w/cond. input from PETR, and generative sampling)
COCO val2017
MogaNet-B (384x288)
MS COCO
OmniPose (WASPv2)
AIC
ITOP front-view
AdaPose
InLoc
GIM-DKM
UPenn Action
OmniPose
MPII Single Person
4xRSN-50
J-HMDB
SimpleBaseline + HANet
ITOP top-view
DECA-D3
SALSA
SubdivNet
300W (Full)
DensePose-COCO
Parsing R-CNN + ResNext101
FLIC Wrists
Stacked Hourglass Networks
BRACE
HRNet fine-tuned on BRACE
UAV-Human
AlphaPose
COCO 2017 val
LOGO-CAP (Ours) HRNet-W48
FLIC Elbows
Stacked Hourglass Networks
ApolloCar3D
KITTI 2015
GeoNet
3DPW
MERL-RAV
SPIGA
Pix3D
Mid-Level based
MS-COCO
UniHCP (finetune)
MPII
OmniPose (WASPv2)
COCO minival
MSPN
!(()&&!|*|*|
Nate