opencv_zoo / benchmark /config /pose_estimation_mediapipe.yaml
Wanli
add pose estimation model (#152)
ab8d410
Benchmark:
name: "Pose Estimation Benchmark"
type: "Recognition"
data:
path: "data/person_detection"
files: ["person1.jpg", "person2.jpg", "person3.jpg"]
sizes: # [[w1, h1], ...], Omit to run at original scale
- [256, 256]
metric:
warmup: 30
repeat: 10
backend: "default"
target: "cpu"
Model:
name: "MPPose"
confThreshold: 0.9