Detectron: how to understand 'skeleton' in coco datasets keypoint?

Created on 30 Aug 2018  ·  2Comments  ·  Source: facebookresearch/Detectron

hi!
In the coco datasets, I use keypoints, and there is "skeleton" in categories such as:

"skeleton": [
                [
                    16, 
                    14
                ], 
                [
                    14, 
                    12
                ], 
                [
                    17, 
                    15
                ], 
                [
                    15, 
                    13
                ], 
                [
                    12, 
                    13
                ], 
                [
                    6, 
                    12
                ], 
                [
                    7, 
                    13
                ], 
                [
                    6, 
                    7
                ], 
                [
                    6, 
                    8
                ], 
                [
                    7, 
                    9
                ], 
                [
                    8, 
                    10
                ], 
                [
                    9, 
                    11
                ], 
                [
                    2, 
                    3
                ], 
                [
                    1, 
                    2
                ], 
                [
                    1, 
                    3
                ], 
                [
                    2, 
                    4
                ], 
                [
                    3, 
                    5
                ], 
                [
                    4, 
                    6
                ], 
                [
                    5, 
                    7
                ]
            ]

In coco document, it declares such as:
"skeleton", which defines connectivity via a list of keypoint edge pairs and is used for visualization.
But I can't understand what it means. Can you help me please? Thanks!

Most helpful comment

They are indexes starting at 1 (NOT ZERO) into the joint lookup table. Example [6,12] is ['left_shoulder', left_hip']

"keypoints": [ "nose", "left_eye", "right_eye", "left_ear", "right_ear", "left_shoulder", "right_shoulder", "left_elbow", "right_elbow", "left_wrist", "right_wrist", "left_hip", "right_hip", "left_knee", "right_knee", "left_ankle", "right_ankle" ]

All 2 comments

They are indexes starting at 1 (NOT ZERO) into the joint lookup table. Example [6,12] is ['left_shoulder', left_hip']

"keypoints": [ "nose", "left_eye", "right_eye", "left_ear", "right_ear", "left_shoulder", "right_shoulder", "left_elbow", "right_elbow", "left_wrist", "right_wrist", "left_hip", "right_hip", "left_knee", "right_knee", "left_ankle", "right_ankle" ]

@teaglin thanks!

Was this page helpful?
0 / 5 - 0 ratings