Home

Awesome

V2V-PoseNet-pytorch

This is a pytorch implementation of V2V-PoseNet(V2V-PoseNet: Voxel-to-Voxel Prediction Network for Accurate 3D Hand and Human Pose Estimation from a Single Depth Map), which is largely based on the author's torch7 implementation.

This repository provides

Requirements

Warning on pytorch0.4.1 cudnn:

May need to disable cudnn for batchnorm, or just only use cuda instead. With cudnn for batchnorm and in float precision, the model cannot train well. My simple experiments show that:

cudnn+float: NOT work(e.g. the loss decreases much slower, and result in a higher loss) 
cudnn+float+(disable batchnorm's cudnn): work(e.g. the loss decreases faster, and result in a lower loss)
cudnn+double: work, but the speed is slow
cuda+(float/double): work, but uses much more memroy

There is a similar issue pointed out by https://github.com/Microsoft/human-pose-estimation.pytorch. As suggested, disable cudnn for batchnorm:

PYTORCH=/path/to/pytorch
for pytorch v0.4.0
sed -i "1194s/torch\.backends\.cudnn\.enabled/False/g" ${PYTORCH}/torch/nn/functional.py
for pytorch v0.4.1
sed -i "1254s/torch\.backends\.cudnn\.enabled/False/g" ${PYTORCH}/torch/nn/functional.py

MSRA hand dataset demo

Usage

git clone https://github.com/dragonbook/V2V-PoseNet-pytorch.git
cd V2V-PoseNet-pytorch

Note, this repository contains a copy of the msra hand centers under ./datasets/msra_center.

PYTHONPATH=./ python ./experiments/msra-subject3/main.py

msra_s3_acc

msra_s3_mean_error

Additional IntegralPose/PoseFix style loss implementation

Replaced V2V-PoseNet's loss with PoseFix's loss(one-hot heatmap loss + L1 coord loss), and it's independently implemented under ./integral-pose directory. Also, configure data_dir and center_dir in ./integral-pose/main.py, and start training. The result shows about 10mm mean error.

integral_loss_s3_acc

integral_loss_mean_error

compare_mean_error

Below is from author's README for reference

V2V-PoseNet: Voxel-to-Voxel Prediction Network for Accurate 3D Hand and Human Pose Estimation from a Single Depth Map

Introduction

This is our project repository for the paper, V2V-PoseNet: Voxel-to-Voxel Prediction Network for Accurate 3D Hand and Human Pose Estimation from a Single Depth Map (CVPR 2018).

We, Team SNU CVLAB, (<i>Gyeongsik Moon, Juyong Chang</i>, and <i>Kyoung Mu Lee</i> of Computer Vision Lab, Seoul National University) are winners of HANDS2017 Challenge on frame-based 3D hand pose estimation.

Please refer to our paper for details.

If you find our work useful in your research or publication, please cite our work:

[1] Moon, Gyeongsik, Ju Yong Chang, and Kyoung Mu Lee. "V2V-PoseNet: Voxel-to-Voxel Prediction Network for Accurate 3D Hand and Human Pose Estimation from a Single Depth Map." <i>CVPR 2018. </i> [arXiv]

@InProceedings{Moon_2018_CVPR_V2V-PoseNet,
author = {Moon, Gyeongsik and Chang, Juyong and Lee, Kyoung Mu},
title = {V2V-PoseNet: Voxel-to-Voxel Prediction Network for Accurate 3D Hand and Human Pose Estimation from a Single Depth Map},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
year = {2018}
}

In this repository, we provide

Model Architecture

V2V-PoseNet

HANDS2017 frame-based 3D hand pose estimation Challenge Results

Challenge_result

Comparison with the previous state-of-the-art methods

Paper_result_hand_graph

Paper_result_hand_table

Paper_result_human_table

About our code

Dependencies

Our code is tested under Ubuntu 14.04 and 16.04 environment with Titan X GPUs (12GB VRAM).

Code

Clone this repository into any place you want. You may follow the example below.

makeReposit = [/the/directory/as/you/wish]
mkdir -p $makeReposit/; cd $makeReposit/
git clone https://github.com/mks0601/V2V-PoseNet_RELEASE.git

To train our model, please run the following command in the src directory:

th rum_me.lua

Dataset

We trained and tested our model on the four 3D hand pose estimation and one 3D human pose estimation datasets.

Results

Here we provide the precomputed centers, estimated 3D coordinates and pre-trained models.

The precomputed centers are obtained by training the hand center estimation network from DeepPrior++ . Each line represents 3D world coordinate of each frame. In case of ICVL, NYU, MSRA dataset, if depth map is not exist or not contain hand, that frame is considered as invalid. In case of ITOP dataset, if 'valid' variable of a certain frame is false, that frame is considered as invalid. All test images are considered as valid.

The 3D coordinates estimated on the ICVL, NYU and MSRA datasets are pixel coordinates and the 3D coordinates estimated on the ITOP datasets are world coordinates. The estimated results are from ensembled model. You can make the results from a single model by downloading the pre-trained model and testing it.

We used awesome-hand-pose-estimation to evaluate the accuracy of the V2V-PoseNet on the ICVL, NYU and MSRA dataset.

Belows are qualitative results. result_1 result_2 result_3 result_4 result_5 result_6