Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields ∗
2017-09-01 01:56
423 查看
只找一个limb,即只考虑a single pair of parts j1 and j2, 此时对应到:带权重二分图的匹配问题,权重就是积分出来的E,目标就是找到累积权重最大的匹配
When it comes to finding the full body pose of multiple
people, determining Z is a K-dimensional matching problem.
This problem is NP Hard [32] and many relaxations
exist. In this work, we add two relaxations to the optimization,
specialized to our domain. First, we choose a minimal
number of edges to obtain a spanning tree skeleton of human
pose rather than using the complete graph, as shown in
Fig. 6c. Second, we further decompose the matching problem
into a set of bipartite matching subproblems and determine
the matching in adjacent tree nodes independently,
as shown in Fig. 6d.
Our optimization scheme over
the tree structure is orders of magnitude faster than the optimization
over the fully connected graph 效率提高了几个数量级
PCKh: 各个part落在groundtruth附近,若在head size之内,认为检测出了。PCK的阈值是max(height,width) of body 的0.1 or 0.2倍。同时还可以分析PCKh-0.5等等精度的准确率
mAP
OKS: object keypoint similarity: It is calculated from scale of the person and the
distance between predicted points and GT points.
It is noteworthy
that our method has lower accuracy than the top-down
methods on people of smaller scales (APM). The reason is
that our method has to deal with a much larger scale range
spanned by all people in the image in one shot. In contrast,
top-down methods can rescale the patch of each detected
area to a larger size and thus suffer less degradation
at smaller scales.
If we use the GT bounding box and a single
person CPM [31], we can achieve a upper-bound for
the top-down approach using CPM, which is 62.7% AP.
If we use the state-of-the-art object detector, Single Shot
MultiBox Detector (SSD)[16], the performance drops 10%.
This comparison indicates the performance of top-down approaches
rely heavily on the person detector就是说自顶向下的方法每次都处理一个人,需要有个bounding box把人框出来(1. person detection 2. CPM或其他单人姿态检测算法),然后对框出来的单个人图像进行尺度调整,到一个合适的图像大小,再进行处理。当bounding box精度不够时,自顶向下的方法误差会很高。
9.
When it comes to finding the full body pose of multiple
people, determining Z is a K-dimensional matching problem.
This problem is NP Hard [32] and many relaxations
exist. In this work, we add two relaxations to the optimization,
specialized to our domain. First, we choose a minimal
number of edges to obtain a spanning tree skeleton of human
pose rather than using the complete graph, as shown in
Fig. 6c. Second, we further decompose the matching problem
into a set of bipartite matching subproblems and determine
the matching in adjacent tree nodes independently,
as shown in Fig. 6d.
Our optimization scheme over
the tree structure is orders of magnitude faster than the optimization
over the fully connected graph 效率提高了几个数量级
PCKh: 各个part落在groundtruth附近,若在head size之内,认为检测出了。PCK的阈值是max(height,width) of body 的0.1 or 0.2倍。同时还可以分析PCKh-0.5等等精度的准确率
mAP
OKS: object keypoint similarity: It is calculated from scale of the person and the
distance between predicted points and GT points.
It is noteworthy
that our method has lower accuracy than the top-down
methods on people of smaller scales (APM). The reason is
that our method has to deal with a much larger scale range
spanned by all people in the image in one shot. In contrast,
top-down methods can rescale the patch of each detected
area to a larger size and thus suffer less degradation
at smaller scales.
If we use the GT bounding box and a single
person CPM [31], we can achieve a upper-bound for
the top-down approach using CPM, which is 62.7% AP.
If we use the state-of-the-art object detector, Single Shot
MultiBox Detector (SSD)[16], the performance drops 10%.
This comparison indicates the performance of top-down approaches
rely heavily on the person detector就是说自顶向下的方法每次都处理一个人,需要有个bounding box把人框出来(1. person detection 2. CPM或其他单人姿态检测算法),然后对框出来的单个人图像进行尺度调整,到一个合适的图像大小,再进行处理。当bounding box精度不够时,自顶向下的方法误差会很高。
9.
相关文章推荐
- 姿态论文整理--02-Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields
- caffe openpose/Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields配置(转)
- 姿态估计论文思路整理 -- Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields
- 行人姿态估计--Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields
- Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields ∗ 实时多人人体姿态估计论文原理讲解
- 论文阅读:《Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields》CVPR 2017
- 论文阅读笔记: 2017 cvpr Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields
- Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields
- [OpenPose翻译] Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields ∗原文翻译(注释版)
- 论文阅读-Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields
- 拜读曹哲的Realtime Multi-Person 2D Pose Estimation using Part Affinity Fileds
- Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields 个人解读
- READING NOTE: Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields
- Code note: Realtime Multi-person 2D Pose estimation using Part Affinity Fields(2)
- Paper Reading:Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields
- 论文阅读:RealTime Multi-Person 2D Pose Estimation using Part Affinity Fields
- Paper reading: Realtime Multi-person 2D Pose estimation using Part Affinity Fields(1)
- Using Potential Fields in a Real-time Strategy Game Scenario (Tutorial)
- Human Pose Matching on mobile — a fun application using Human Pose Estimation (Part 1 Intro)
- 论文阅读:《Towards accurate multi-person pose estimation in the wild》CVPR 2017