OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Developer Tools & SDKs

Hyperpose

HyperPose is a Python library for developing custom human pose-estimation models and a C++ library for high-performance model inference. Its inference engine describes real-time CPU/GPU execution using pipeline parallelism, TensorRT inference, and hybrid CPU/GPU scheduling.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Hyperpose — OpenViST project profile in AI Developer Tools & SDKs.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use HyperPose when you need developer-level control over pose-estimation models or want to build a video or camera inference pipeline using its documented C++ container.

Who it is for

  • Developers building custom human pose-estimation applications.
  • Machine-learning practitioners customizing pose models, datasets, or training and evaluation stages.
  • Engineers processing video or camera input with a GPU-backed inference pipeline.

Documented uses

  • Customize pose-estimation architectures and datasets.
  • Configure training, evaluation, visualization, and pre- or post-processing.
  • Run pose inference on video or camera input using the provided Docker image.
Before you choose

Requirements & limits

  • The project separates Python model development from C++ inference.
  • The Docker inference quick start requires NVIDIA GPU container support and a qualifying CUDA driver.
  • The README lists OpenPose, Pifpaf, and PoseProposal Network as example customizable architectures.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Run inference on a video with the Docker image

Use the documented container example to process a video and copy its output file locally.

Before you start

  • Docker CE Engine 19.03 or later.
  • NVIDIA Docker 2.0 or later.
  • CUDA Driver 418.81.07 or later for the default CUDA 10.0 image.
  • The example uses GPU access.

Steps

  1. Pull the image with the command below.

    Documented command
    docker pull tensorlayer/hyperpose

  2. Start the container with the command below.

    Documented command
    docker run --name quick-start --gpus all tensorlayer/hyperpose --runtime=stream

  3. Copy the generated file with the command below.

    Documented command
    docker cp quick-start:/hyperpose/build/output.avi .

  4. Remove the container with the command below.

    Documented command
    docker rm quick-start

Expected outcome

The example copies `output.avi` from the container to the current local directory.

Guide limitations

  • The repository says its example commands were tested on Ubuntu 18.04.
  • This recipe follows the documented image example; it does not establish compatibility with other environments.
Documentation 1 ↗
COMPARE OPTIONS

Compare options for Model inference and serving

These projects document a common task and share a project role. This is a source-based comparison, not an execution benchmark or a guarantee of interchangeability.

ProjectDocumented purposePlatformAccess and requirements
HyperposeCustomize pose-estimation architectures and datasets.Docker, Linux, macOS, Windows
Multi Task NlpDefine multiple tasks and their data and model settings in a YAML file.Linux, macOS, WindowsThe project is an open-source library under the Apache-2.0 license, distributed through its public GitHub repository.
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
18 October 2026
Original website title
tensorlayer/HyperPose · Official homepage title · retained in its source language
How we review sourcesSuggest a correction