GitPlanet

Cheap and reliable Node.js hosting starts at $3/month, and $1/month static HTML hosting

Created with love in Canada, visit hostnodejs.com today

Feel like to post an Ad? Learn Details
All Projects → anibali → eval-mpii-pose

anibali / eval-mpii-pose

Licence: other
Evaluation code for the MPII human pose dataset

Programming Languages

matlab
3953 projects

Labels

mpii-datasetpose-estimation

Projects that are alternatives of or similar to eval-mpii-pose

Fast Stacked Hourglass Network OpenVino
A fast stacked hourglass network for human pose estimation on OpenVino
Stars: ✭ 52 (-10.34%)
Mutual labels:  mpii-dataset, pose-estimation
pytorch-PyraNet
Pytorch version reinplement code of PyraNet , for paper : Learning Feature Pyramids for Human Pose Estimation
Stars: ✭ 32 (-44.83%)
Mutual labels:  mpii-dataset, pose-estimation
GaitGraph
Official repository for "GaitGraph: Graph Convolutional Network for Skeleton-Based Gait Recognition" (ICIP'21)
Stars: ✭ 68 (+17.24%)
Mutual labels:  pose-estimation
Keypoint Communities
[ICCV '21] In this repository you find the code to our paper "Keypoint Communities".
Stars: ✭ 255 (+339.66%)
Mutual labels:  pose-estimation
terran
A human perception library
Stars: ✭ 98 (+68.97%)
Mutual labels:  pose-estimation
latent-pose-reenactment
The authors' implementation of the "Neural Head Reenactment with Latent Pose Descriptors" (CVPR 2020) paper.
Stars: ✭ 132 (+127.59%)
Mutual labels:  pose-estimation
GA SLAM
🚀 SLAM for autonomous planetary rovers with global localization
Stars: ✭ 40 (-31.03%)
Mutual labels:  pose-estimation
Openpose-based-GUI-for-Realtime-Pose-Estimate-and-Action-Recognition
GUI based on the python api of openpose in windows using cuda10 and cudnn7. Support body , hand, face keypoints estimation and data saving. Realtime gesture recognition is realized through two-layer neural network based on the skeleton collected from the gui.
Stars: ✭ 69 (+18.97%)
Mutual labels:  pose-estimation
slamkit
SLAM Kit
Stars: ✭ 28 (-51.72%)
Mutual labels:  pose-estimation
MSPN
Multi-Stage Pose Network
Stars: ✭ 40 (-31.03%)
Mutual labels:  pose-estimation
kapao
KAPAO is an efficient single-stage human pose estimation model that detects keypoints and poses as objects and fuses the detections to predict human poses.
Stars: ✭ 604 (+941.38%)
Mutual labels:  pose-estimation
posture recognition
Posture recognition based on common camera
Stars: ✭ 91 (+56.9%)
Mutual labels:  pose-estimation
DREAM
DREAM: Deep Robot-to-Camera Extrinsics for Articulated Manipulators (ICRA 2020)
Stars: ✭ 109 (+87.93%)
Mutual labels:  pose-estimation
FastPose
pytorch realtime multi person keypoint estimation
Stars: ✭ 36 (-37.93%)
Mutual labels:  pose-estimation
PCLoc
Pose Correction for Highly Accurate Visual Localization in Large-scale Indoor Spaces (ICCV 2021)
Stars: ✭ 37 (-36.21%)
Mutual labels:  pose-estimation
ailia-models
The collection of pre-trained, state-of-the-art AI models for ailia SDK
Stars: ✭ 1,102 (+1800%)
Mutual labels:  pose-estimation
poseviz
3D Human Pose Visualizer for Python
Stars: ✭ 68 (+17.24%)
Mutual labels:  pose-estimation
AirPose
This repository contains the code of AirPose, our multi-view fusion network for Human Pose and Shape Estimation method
Stars: ✭ 39 (-32.76%)
Mutual labels:  pose-estimation
mediapipe-osc
MediaPipe examples which stream their detections over OSC.
Stars: ✭ 26 (-55.17%)
Mutual labels:  pose-estimation
Everything-Will-Be-IK
A Robust Inverse Kinematics Library
Stars: ✭ 35 (-39.66%)
Mutual labels:  pose-estimation
View All Similar Projects ➔

eval-mpii-pose

Scripts for evaluating results on the MPII human pose dataset. "Single person pose estimation" is the only type of evaluation currently supported.

Disclaimer: This is an unofficial repository, I am not from MPI and I was not involved in the creation of the dataset.

Input format

Predictions are expected to have the following format:

  • Must be a Matlab (.mat) or HDF5 (.h5) file
    • Must have one field, preds, which is the joint predictions tensor
    • Tensor size must be [2 x 16 x n] or [n x 16 x 2]
  • Must correspond to one of the following subsets: train, val, test
    • See annot/{train,valid,test}.h5 for which examples are in each of these subsets

Predictions produced by the following repositories meet these requirements:

Metrics

PCKh

The PCKh performance metric is the percentage of joints with predicted locations that are no further than half of the head segment length from the ground truth.

"PCKh total" excludes the pelvis and thorax joints from the calculation, presumably because they are very easy to predict given that the approximate person center is provided.

Scripts

evalMPII.m

Loads predictions from Matlab or HDF5 files and compares them with ground truth labels to calculate accuracy metrics (eg PCKh). You will want to edit getExpParamsNew.m to add new sets of predictions, and evalMPII.m to specify which predictions to include and which subset (train/val) to use.

prepareTestResults.m

Loads flat test set predictions and prepares them for submission.

Reference predictions

The preds/reference directory contains multiple validation set prediction files generated by established pose estimation models. You can compare against these predictions using evalMPII.m.

NOTE: Since the reference predictions are for the validation set, they are not compatible with the prepareTestResults.m script.

File origins

In order to keep evaluation in line with existing work, a lot of files in this repository were copied verbatim from other sources.

Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].