OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Models

Cnngesturerecognizer

CNNGestureRecognizer is a Python computer-vision project that uses OpenCV to capture and preprocess hand images and a Keras convolutional neural network to classify gestures.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Cnngesturerecognizer — OpenViST project profile in AI Models.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it as a local starting point when you want to examine or adapt a camera-based gesture-recognition pipeline rather than call a hosted recognition service.

Who it is for

  • Developers prototyping camera-based gesture interfaces.
  • Computer-vision learners exploring CNN training and feature-map visualization.
  • Python users adapting gesture labels or model architecture for a local experiment.

Documented uses

  • Predict one of five pretrained gesture labels from camera input.
  • Save prediction output to the console or a JSON file.
  • Create gesture image samples and retrain the model.
  • Visualize feature maps for an input image.
Before you choose

Requirements & limits

The repository is public and specifies an MIT license. The README lists Python 3.6.1, OpenCV 3.4.1 and Keras 2.0.2; it notes TensorFlow support and describes Theano as obsolete and unsupported.

  • The pretrained labels are OK, PEACE, STOP, PUNCH and NOTHING.
  • Binary preprocessing is described for an empty, plain background; SkinMask is described for good lighting when the background is not empty.
  • The macOS and Windows pretrained weight files are listed separately, and the macOS file is hosted on Google Drive.
  • The README lists older dependency versions, so users should review setup requirements before choosing it for a new environment.

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 the gesture-recognition launcher on Windows

Launch the project using its documented TensorFlow backend setting.

Before you start

  • Python 3.6.1
  • OpenCV 3.4.1
  • Keras 2.0.2
  • TensorFlow 1.2.1
  • The project files, including trackgesture.py

Steps

  1. Set the KERAS_BACKEND environment variable to tensorflow: set "KERAS_BACKEND=tensorflow".

  2. Run the launcher with Python: python trackgesture.py.

Expected outcome

The project launcher starts with TensorFlow selected as the Keras backend; the application provides prediction, training, and visualization functions.

Guide limitations

  • The README lists specific, old dependency versions; Theano 0.9.0 is identified as obsolete and unsupported.
  • The README's Mac command and its accompanying backend description conflict, so this recipe is limited to the Windows instructions.
  • Pretrained gesture weights are described as OS-specific, and no Windows weight-download or setup procedure is documented.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
13 October 2026
Original website title
asingh33/CNNGestureRecognizer · Official homepage title · retained in its source language
How we review sourcesSuggest a correction