Direct Multi-view Multi-person 3D Human Pose Estimation

Related tags

Miscellaneousmvp
Overview

Implementation of NeurIPS-2021 paper: Direct Multi-view Multi-person 3D Human Pose Estimation

[paper] [video-YouTube, video-Bilibili] [slides]

This is the official implementation of our NeurIPS-2021 work: Multi-view Pose Transformer (MvP). MvP is a simple algorithm that directly regresses multi-person 3D human pose from multi-view images.

Framework

mvp_framework

Example Result

mvp_framework

Reference

@article{wang2021mvp,
  title={Direct Multi-view Multi-person 3D Human Pose Estimation},
  author={Tao Wang and Jianfeng Zhang and Yujun Cai and Shuicheng Yan and Jiashi Feng},
  journal={Advances in Neural Information Processing Systems},
  year={2021}
}

1. Installation

  1. Set the project root directory as ${POSE_ROOT}.
  2. Install all the required python packages (with requirements.txt).
  3. compile deformable operation for projective attention.
cd ./models/ops
sh ./make.sh

2. Data and Pre-trained Model Preparation

2.1 CMU Panoptic

Please follow VoxelPose to download the CMU Panoptic Dataset and PoseResNet-50 pre-trained model.

The directory tree should look like this:

${POSE_ROOT}
|-- models
|   |-- pose_resnet50_panoptic.pth.tar
|-- data
|   |-- panoptic
|   |   |-- 16060224_haggling1
|   |   |   |-- hdImgs
|   |   |   |-- hdvideos
|   |   |   |-- hdPose3d_stage1_coco19
|   |   |   |-- calibration_160224_haggling1.json
|   |   |-- 160226_haggling1
|   |   |-- ...

2.2 Shelf/Campus

Please follow VoxelPose to download the Shelf/Campus Dataset.

Due to the limited and incomplete annotations of the two datasets, we use psudo ground truth 3D pose generated from VoxelPose to train the model, we expect mvp would perform much better with absolute ground truth pose data.

Please use voxelpose or other methods to generate psudo ground truth for the training set, you can also use our generated psudo GT: psudo_gt_shelf. psudo_gt_campus. psudo_gt_campus_fix_gtmorethanpred.

Due to the small dataset size, we fine-tune Panoptic pre-trained model to Shelf and Campus. Download the pretrained MvP on Panoptic from model_best_5view and model_best_3view_horizontal_view or model_best_3view_2horizon_1lookdown

The directory tree should look like this:

${POSE_ROOT}
|-- models
|   |-- model_best_5view.pth.tar
|   |-- model_best_3view_horizontal_view.pth.tar
|   |-- model_best_3view_2horizon_1lookdown.pth.tar
|-- data
|   |-- Shelf
|   |   |-- Camera0
|   |   |-- ...
|   |   |-- Camera4
|   |   |-- actorsGT.mat
|   |   |-- calibration_shelf.json
|   |   |-- pesudo_gt
|   |   |   |-- voxelpose_pesudo_gt_shelf.pickle
|   |-- CampusSeq1
|   |   |-- Camera0
|   |   |-- Camera1
|   |   |-- Camera2
|   |   |-- actorsGT.mat
|   |   |-- calibration_campus.json
|   |   |-- pesudo_gt
|   |   |   |-- voxelpose_pesudo_gt_campus.pickle
|   |   |   |-- voxelpose_pesudo_gt_campus_fix_gtmorethanpred_case.pickle

2.3 Human3.6M dataset

Please follow CHUNYUWANG/H36M-Toolbox to prepare the data.

2.4 Full Directory Tree

The data and pre-trained model directory tree should look like this, you can only download the Panoptic dataset and PoseResNet-50 for reproducing the main MvP result and ablation studies:

${POSE_ROOT}
|-- models
|   |-- pose_resnet50_panoptic.pth.tar
|   |-- model_best_5view.pth.tar
|   |-- model_best_3view_horizontal_view.pth.tar
|   |-- model_best_3view_2horizon_1lookdown.pth.tar
|-- data
|   |-- pesudo_gt
|   |   |-- voxelpose_pesudo_gt_shelf.pickle
|   |   |-- voxelpose_pesudo_gt_campus.pickle
|   |   |-- voxelpose_pesudo_gt_campus_fix_gtmorethanpred_case.pickle
|   |-- panoptic
|   |   |-- 16060224_haggling1
|   |   |   |-- hdImgs
|   |   |   |-- hdvideos
|   |   |   |-- hdPose3d_stage1_coco19
|   |   |   |-- calibration_160224_haggling1.json
|   |   |-- 160226_haggling1
|   |   |-- ...
|   |-- Shelf
|   |   |-- Camera0
|   |   |-- ...
|   |   |-- Camera4
|   |   |-- actorsGT.mat
|   |   |-- calibration_shelf.json
|   |   |-- pesudo_gt
|   |   |   |-- voxelpose_pesudo_gt_shelf.pickle
|   |-- CampusSeq1
|   |   |-- Camera0
|   |   |-- Camera1
|   |   |-- Camera2
|   |   |-- actorsGT.mat
|   |   |-- calibration_campus.json
|   |   |-- pesudo_gt
|   |   |   |-- voxelpose_pesudo_gt_campus.pickle
|   |   |   |-- voxelpose_pesudo_gt_campus_fix_gtmorethanpred_case.pickle
|   |-- HM36

3. Training and Evaluation

The evaluation result will be printed after every epoch, the best result can be found in the log.

3.1 CMU Panoptic dataset

We train and validate on the five selected camera views. We trained our models on 8 GPUs and batch_size=1 for each GPU, note the total iteration per epoch should be 3205, if not, please check your data.

python -m torch.distributed.launch --nproc_per_node=8 --use_env run/train_3d.py --cfg configs/panoptic/best_model_config.yaml

Pre-trained models

Datasets AP25 AP25 AP25 AP25 MPJPE pth
Panoptic 92.3 96.6 97.5 97.7 15.8 here

3.1.1 Ablation Experiments

You can find several ablation experiment configs under ./configs/panoptic/, for example, removing RayConv:

python -m torch.distributed.launch --nproc_per_node=8 --use_env run/train_3d.py --cfg configs/panoptic/ablation_remove_rayconv.yaml

3.2 Shelf/Campus datasets

As shelf/campus are very small dataset with incomplete annotation, we finetune pretrained MvP with pseudo ground truth 3D pose extracted with VoxelPose, we expect more accurate GT would help MvP achieve much higher performance.

python -m torch.distributed.launch --nproc_per_node=8 --use_env run/train_3d.py --cfg configs/shelf/mvp_shelf.yaml

Pre-trained models

Datasets Actor 1 Actor 2 Actor 2 Average pth
Shelf 99.3 95.1 97.8 97.4 here
Campus 98.2 94.1 97.4 96.6 here

3.3 Human3.6M dataset

MvP also applies to the naive single-person setting, with dataset like Human3.6, to come

python -m torch.distributed.launch --nproc_per_node=8 --use_env run/train_3d.py --cfg configs/h36m/mvp_h36m.yaml

4. Evaluation Only

To evaluate a trained model, pass the config and model pth:

python -m torch.distributed.launch --nproc_per_node=8 --use_env run/validate_3d.py --cfg xxx --model_path xxx

LICENSE

This repo is under the Apache-2.0 license. For commercial use, please contact the authors.

Owner
Sea AI Lab
Sea AI Lab
Emulate and Dissect MSF and *other* attacks

Need help in analyzing Windows shellcode or attack coming from Metasploit Framework or Cobalt Strike (or may be also other malicious or obfuscated code)? Do you need to automate tasks with simple scr

123 Dec 16, 2022
x-tools is a collection of tools developed in Python

x-tools X-tools is a collection of tools developed in Python Commands\

5 Jan 24, 2022
TikTok Auto Claimer Made By Aim low!#9999 Leaked By bazooka#0001

Zues Auto Claimer Leaked By bazooka#0001 put proxies in prox.txt put ssid in sid.txt put all users you want to target in user.txt for the login just t

1 Jan 14, 2022
Python script to automate the change of desktop background

wallomator Python script to automate the change of desktop background A python script that automates the process of changing the desktop background. I

Mohammed Haaris Javed 10 Jun 16, 2022
python scripts and other files to generate induction encoder PCBs in Kicad

induction_encoder python scripts and other files to generate induction encoder PCBs in Kicad Targeting the Renesas IPS2200 encoder chips.

Taylor Alexander 8 Feb 16, 2022
This is some simple code to scrape vistbook's system to get an overview of the different cabins availability.

DNT_cabin_availability_system This is some simple code to scrape visbook's system to get an overview of the different cabins availability. The system

Andreas Lorentzen 1 Sep 25, 2022
GitHub Actions Version Updater Updates All GitHub Action Versions in a Repository and Creates a Pull Request with the Changes.

GitHub Actions Version Updater GitHub Actions Version Updater is GitHub Action that is used to update other GitHub Actions in a Repository and create

Maksudul Haque 42 Dec 22, 2022
Null safe support for Python

Null Safe Python Null safe support for Python. Installation pip install nullsafe Quick Start Dummy Class class Dummy: pass Normal Python code: o =

Paaksing 13 Nov 17, 2022
An ongoing curated list of frameworks, libraries, learning tutorials, software and resources in Python Language.

Python Development Welcome to the world of Python. An ongoing curated list of frameworks, libraries, learning tutorials, software and resources in Pyt

Paul Veillard 2 Dec 24, 2021
My attempt at this years Advent of Code!

Advent-of-code-2021 My attempt at this years Advent of Code! day 1: ** day 2: ** day 3: ** day 4: ** day 5: ** day 6: ** day 7: ** day 8: * day 9: day

1 Jul 06, 2022
Script to use SysWhispers2 direct system calls from Cobalt Strike BOFs

SysWhispers2BOF Script to use SysWhispers2 direct system calls from Cobalt Strike BOFs. Introduction This script was initially created to fix specific

FalconForce 101 Dec 20, 2022
ClamNotif: A tool to send you ClamAV notifications

A tool to forward notifications to different recipients categorised by two severity levels of the regular health reports produced by `clamscan` bundled with the ClamAV antivirus engine.

PiSoft Company Ltd. 1 Nov 15, 2021
A simple python script that print the Mandelbrot set for every power of the formal formula.

Python Mandelbrot A simple python script that print the Mandelbrot set for every power of the formal formula.

Paride Giunta 2 Apr 15, 2022
Let's make a lot of random function from Scracth...

Pseudo-Random On a whim I asked myself the question about how randomness is integrated into an algorithm? So I started the adventure by trying to code

Yacine 2 Jan 19, 2022
Python-Kite: Simple python code to make kite pattern

Python-Kite Simple python code to make kite pattern. Getting Started These instr

Anoint 0 Mar 22, 2022
3x+1 recreated in Python

3x-1 3x+1 recreated in Python If a number is odd it is multiplied by 3 and 1 is added to the product. If a number is even it is divided by 2. These ru

4 Aug 19, 2022
🦋 hundun is a python library for the exploration of chaos.

hundun hundun is a python library for the exploration of chaos. Please note that this library is in beta phase. Example Import the package's equation

kosh 7 Nov 07, 2022
A simple app that helps to train quick calculations.

qtcounter A simple app that helps to train quick calculations. Usage Manual Clone the repo in a folder using git clone https://github.com/Froloket64/q

0 Nov 27, 2021
Terminal compatible with ansi-bbs. Meant to be a prototype, but published because why not.

pybbsterm: Terminal emulator for calling BBSs. Use cases (non-exhaustive) Explore terminal protocols. Connect to BBSs. Highlights Python 3.8+ code. Bu

Roc Vallès i Domènech 9 Apr 29, 2022
Introduction to Databases Coursework 2 (SQL) - dataset generator

Introduction to Databases Coursework 2 (SQL) - dataset generator This is python script generates a text file with insert queries for the schema.sql fi

Javier Bosch 1 Nov 08, 2021