GithubHelp home page GithubHelp logo

summerwings / human_object_interaction Goto Github PK

View Code? Open in Web Editor NEW

This project forked from romerobarata/human_object_interaction

0.0 0.0 0.0 65 KB

Code for CVPR'21 paper "Learning Asynchronous and Sparse Human-Object Interaction in Videos".

Python 100.00%

human_object_interaction's Introduction

Video Human-Object Interaction

Code for the CVPR'21 paper Learning Asynchronous and Sparse Human-Object Interaction in Videos.

Environment Setup

First please create an appropriate environment using conda:

conda env create -f environment.yml

conda activate vhoi

Download Data and Pre-Trained Models

Please download the necessary data for the CAD-120 and Bimanual Actions datasets from the link below, and put the downloaded data folder in this current directory (i.e. ./data/...).

Link: data.

Pre-trained models can be found in the link below, and the outputs folder should be placed in this current directory as well (i.e. ./outputs/...).

Link: models.

Test Pre-Trained Models

Evaluate ASSIGN on CAD-120 dataset:

python -W ignore predict.py --pretrained_model_dir ./outputs/cad120/assign/hs512_e40_bs16_lr0.001_sc-None_h2h-False_h2o-True_o2h-True_o2o-True_m-v2-v1-att-v3-False-True_sd-0.1-True_os-ind_dn-1-gs_pf-e0s0_c0_sp-0_ihs-False_ios-False_bl-False-1.0-1.0_sl-True-False-4.0-1.0_fl0-0.0_mt-False_pt-True-z_gc0.0_ds3_Subject1 --cross_validate

Evaluate ASSIGN on Bimanual Actions dataset:

python -W ignore predict.py --pretrained_model_dir ./outputs/bimanual/assign/hs64_e30_bs32_lr0.001_sc-None_h2h-True_h2o-True_o2h-True_o2o-True_m-v2-v1-att-v3-False-True_sd-0.1-True_os-ind_dn-1-gs_pf-e0s0_c0_sp-0_ihs-False_ios-False_bl-False-1.0-1.0_sl-True-False-4.0-1.0_fl0-0.0_mt-False_pt-True-z_gc0.0_ds3_1 --cross_validate

Train a Model

To train a model from scratch, edit the ./conf/config.yaml file, and depending on the selected dataset and model, also edit the associated model .yaml file in ./conf/models/ and the associated dataset .yaml file in ./conf/data/. After editing the files, just run python train.py.

The configuration settings used for the provided pre-trained models can be found inside the pre-trained model directory, within the hidden .hydra folder. For example, ./outputs/cad120/assign/hs512_e40_bs16_lr0.001_sc-None_h2h-False_h2o-True_o2h-True_o2o-True_m-v2-v1-att-v3-False-True_sd-0.1-True_os-ind_dn-1-gs_pf-e0s0_c0_sp-0_ihs-False_ios-False_bl-False-1.0-1.0_sl-True-False-4.0-1.0_fl0-0.0_mt-False_pt-True-z_gc0.0_ds3_Subject1/.hydra/config.yaml.

human_object_interaction's People

Contributors

romerobarata avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.