Computer Vision
Showing 36 of 630 projects
A deep learning-based, threshold-agnostic, subpixel-accurate 2D and 3D spot detection method for fluorescence microscopy and spatial transcriptomics.
ROS driver for Point Grey cameras using the official FlyCapture2 SDK, enabling HDR and physics-based vision.
ROS wrapper for Orbbec Astra 3D cameras, enabling depth sensing and point cloud generation in ROS Kinetic, Melodic, and Noetic.
A TensorFlow implementation of the Mnemonic Descent Method for end-to-end face alignment.
An open-source benchmark solution for the Kaggle TGS Salt Identification Challenge using semantic segmentation.
Flax implementations and pretrained checkpoints for ResNet, Wide ResNet, ResNeXt, ResNet-D, and ResNeSt in JAX.
A Python devkit for working with the Boreas and Boreas Road Trip all-weather autonomous driving datasets.
A freely usable dataset of over 5,000 labeled clothing images across 20 categories for machine learning projects.
A NativeScript plugin for building augmented reality experiences on iOS and Android.
A 3D object detection method that exploits visibility information from LiDAR point clouds to improve accuracy.
A ROS package for using CSI cameras on Nvidia Jetson platforms (TK1/TX1/TX2) with ROS via gstreamer and Nvidia multimedia API.
Library and utilities for working with ifm pmd-based 3D Time-of-Flight cameras, supporting O3R, O3D, and O3X platforms.
A deep learning model for classifying image aesthetic quality using Inception modules and fine-tuned connected layers.
A native extension for Adobe AIR that enables developers to integrate Microsoft Kinect motion sensing capabilities into desktop applications.
A Python package providing popular computer vision model architectures built with Equinox for JAX.
An application that uses IBM Watson AI services and Cloud Functions to analyze videos, extracting visual and audio insights for search and categorization.
A ROS2 wrapper for real-time object detection, 3D localization, and tracking using RGB-D camera inputs.
A tool for automatically detecting and suggesting mitigation for object, attribute, and geography-based biases in visual datasets.
A tutorial and demo using Hyperopt to auto-optimize CNN architecture and hyperparameters for the CIFAR-100 dataset with Keras/TensorFlow.
A ROS2 node for capturing video from USB cameras using OpenCV, publishing image topics and camera info.
Official JAX implementation of XMC-GAN for text-to-image generation using cross-modal contrastive learning.
C++ implementation of Local Binary Pattern texture descriptors with OpenCV and FFTW3 integration for fast computation.
Build persistent cloud-based AR apps that anchor digital content to real-world locations using ARKit and Swift.
A city-scale dataset and platform for learning holistic 3D structures from panoramic and perspective imagery with detailed annotations.
A high-precision, grid-based C++ library for ground segmentation in LiDAR point clouds, designed for safety-critical autonomous driving and robotics.
A Vue 3 component library for scanning QR codes from camera streams, image captures, and file uploads.
A graphical user interface for annotating point clouds and 3D scenes with bounding boxes, keypoints, and rectangles.
A collection of refactored, high-quality Android examples demonstrating TensorFlow Lite for on-device machine learning tasks.
A ROS2 package that accelerates training and deployment of computer vision models for industrial applications.
A TensorLayer re-implementation of CycleGAN with improvements like resize-convolution and instance normalization.
A Wagtail extension that automatically generates image descriptions and tags using computer vision services.
A curated archive of pre-trained computer vision models for object detection, face recognition, fire detection, and more.
A macOS app that automatically extracts and annotates facial landmarks from videos for GAN training datasets.
A TensorFlow-based model that generates descriptive captions for images using an Inception-v3 encoder and LSTM decoder.
A UE4 plugin that captures RGB and depth data from the Unreal world and publishes it to ROS.
![[Web]](/_next/image?url=https%3A%2F%2Fgithub.com%2Flongjon.png%3Fsize%3D40&w=96&q=75){kind=small}
A pre-release Caffe branch for fully convolutional networks (FCNs), now deprecated with features merged into Caffe master.
Related Tags
Found a gem we're missing?
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.