Open-source alternatives to Camera Samples

TakePhoto is an Android media capture framework and image processing library designed to manage photo acquisition and manipulation. It provides a toolkit for capturing new images via the device camera and selecting existing images from the system gallery or local file storage. The framework distinguishes itself through automatic activity and session state restoration, which recovers the capture process after the system recycles an activity. It also includes a unified handler for managing runtime media permissions for camera and storage access across different Android versions. The library co

RT-DETR is a real-time object detection model based on the detection transformer architecture. It is implemented as a computer vision model for both the PyTorch and PaddlePaddle deep learning platforms, designed to identify and locate multiple objects in images and video streams. The model eliminates the need for anchor generation and non-maximum suppression by utilizing a transformer-based approach. It focuses on high-performance detection, balancing precision and low latency for live environment deployment. The system employs a hybrid encoder and multi-scale feature fusion to extract globa

NanoDet-Plus⚡Super fast and lightweight anchor-free object detection model. 🔥Only 980 KB(int8) / 1.8MB (fp16) and run 97FPS on cellphone🔥

This project is a cross-platform mobile camera framework and real-time computer vision library. It provides a high-performance interface for mobile applications to handle hardware control, media capture, and live camera frame processing. The framework includes a dedicated system for running AI models and custom analysis on live camera streams using high-performance worklets. It also functions as a real-time detection and decoding system for QR codes and barcodes. Broad capabilities cover the capture of high-resolution photos and videos with controls for zoom, HDR, and frame rates. The projec

YOLOv10 is a PyTorch computer vision library and real-time vision framework designed for locating and identifying multiple objects in images and video streams. It functions as an end-to-end object detector that optimizes for high-speed deployment and detection precision. The project is distinguished by an NMS-free detection architecture that predicts a single bounding box per object, eliminating the need for non-maximum suppression post-processing to reduce inference latency. It further optimizes for edge hardware through scalable weights and a quantization-friendly structure that facilitates

This is a ZXing-based toolkit for integrating barcode and QR code scanning and generation into Android applications. It functions as a barcode scanner capable of decoding data from both live camera streams and static image files. The library includes a QR code generator that supports the creation of image files with custom logos embedded in the center of the code. It also provides a customizable camera scanning interface, allowing for adjustments to the scanning overlay and control over the device flash to improve visibility in low light. Its broader capabilities cover the extraction of text

TengineKit is a mobile computer vision software development kit designed for real-time inference on local hardware. It functions as a neural network engine that executes deep learning models directly on mobile devices, enabling applications to perform complex visual analysis without relying on cloud connectivity. The framework provides specialized tools for detecting and tracking human features, including faces, hands, bodies, and irises, alongside general object detection capabilities. By utilizing a native core runtime and hardware-accelerated execution, the library processes visual data lo

FairMOT is a multi-object tracking framework and deep learning model designed to identify and track multiple entities across video frames. It implements a unified pipeline that integrates object detection and identity re-identification into a single-stage joint network. The system utilizes an anchor-free detection method to predict object centers and bounding box dimensions. It maintains identity consistency across consecutive frames by generating high-dimensional embedding vectors for re-identification and employing a Kalman filter for motion state prediction. The framework covers a broad r

labelImg is a desktop image annotation tool and dataset preparation utility used to create labeled datasets for computer vision training. It provides a graphical interface for drawing bounding boxes around objects in images and assigning them class labels to build ground truth data for machine learning models. The software specifically supports the Pascal VOC XML annotation format, exporting image coordinates and class names into standard XML or text structures. It allows users to load predefined class lists from text files to standardize naming across an entire project. Beyond initial label

This project is a PyTorch implementation of the YOLOv4 object detection framework. It provides a system for training and deploying neural networks that identify and locate multiple objects within images and video streams. The framework includes tools for converting trained weights into universal formats and hardware-specific optimized engines, specifically supporting ONNX and TensorRT. It features a TensorRT inference optimizer to reduce latency and increase throughput, as well as a model architecture compatible with NVIDIA DeepStream streaming analytics pipelines. The system covers model tr

YOLOv7 is a PyTorch vision library and real-time inference engine designed for object detection, human pose estimation, and instance segmentation. It provides a framework for detecting and locating multiple objects within images or video streams using neural networks. The system includes tools for custom model training and fine-tuning, allowing pre-trained weights to be adapted to specialized datasets via transfer learning. It also supports model weight export and format conversion to facilitate deployment on production servers and embedded edge devices.

Cameraview is an Android camera integration library used to embed live camera previews and capture photos or videos within applications. It provides a reusable UI component that manages camera hardware initialization and real-time stream rendering. The library includes a media capture framework to control hardware settings such as focus, flash, and aspect ratios for recording images and video files. The project covers functional domains including Android hardware configuration via layout attributes and the rendering of continuous camera streams for immediate visual feedback.

Grafika is a collection of reference implementations and benchmarking tools for the Android platform. It provides technical demonstrations for validating camera lifecycles, GPU performance, and media codec configurations, including a suite of OpenGL ES examples and a graphics reference application. The project distinguishes itself through specialized tools for measuring graphics performance, such as benchmarks for texture upload speeds and pixel read latency. It also implements specific hardware-focused patterns, such as using circular buffers for continuous video capture and utilizing virtua

ImagePicker is a mobile image picker component that provides a unified interface for selecting photos from a gallery or capturing new images via the device camera. It serves as a branded UI media selector and a device camera integration layer. The system operates as a memory-optimized media gallery, utilizing asset identifiers instead of raw binary data to prevent application memory overload. It allows for the customization of the interface, including colors, fonts, and text labels, through a configuration object to match specific brand identities. The component includes capabilities for man

react-native-image-picker is a cross-platform mobile media picker used to select images and videos from a system gallery or capture new media via the device camera. It serves as a native device camera interface and a mobile gallery file picker for accessing local storage. The library includes a media metadata extractor to retrieve technical details from selected files, such as image dimensions, file size, video duration, and EXIF data. The project provides a bridge between JavaScript and native mobile operating systems to trigger system-provided gallery and camera interfaces. It handles the

YOLOX is a high-performance anchor-free YOLO, exceeding yolov3~v5 with MegEngine, ONNX, TensorRT, ncnn, and OpenVINO supported. Documentation: https://yolox.readthedocs.io/

This project is a cross-platform mobile media picker that provides a native interface for selecting images and videos from a device gallery or capturing them via the camera. It acts as a bridge for mobile camera integration and a native gallery selector to import visual assets into a JavaScript environment. The library covers media library access and the handling of user content uploads by allowing users to provide image or video files from their device. This includes the ability to launch the device camera for photos and videos as well as selecting multiple media files from the local library

ZLPhotoBrowser is a mobile image picker library that provides a media gallery interface for selecting photos, videos, and GIFs from a device library. It implements interaction patterns common to social messaging platforms and includes localized support for global users. The library features an integrated camera workflow for capturing new photos and videos directly within the selection process. It also includes an in-app media editor with tools for cropping, drawing, and applying filters to content before it is finalized. The project covers media gallery management, allowing users to organize

This project is a PyTorch implementation of the YOLOv3 object detection architecture. It functions as a real-time object detector and computer vision framework designed to identify and locate multiple objects within images using bounding boxes and class labels. The system allows for both the use of pretrained weights for immediate image analysis and the training of custom models using datasets with bounding box annotations. It provides a programmatic interface to integrate detection capabilities directly into other software applications. The framework includes tools for model evaluation to m

This application is a real-time computer vision system designed to identify and label objects within live video feeds, recorded files, and static images. It functions as a comprehensive framework that integrates pre-trained machine learning models with video processing pipelines to perform multi-object localization and visual data tracking. The system distinguishes itself through a multithreaded architecture that decouples frame acquisition from detection logic, ensuring the interface remains responsive during continuous analysis. It provides specialized scripts for training and optimizing cu

PictureSelector is an Android media selection library and toolkit for browsing and picking images, videos, and audio files from a device album. It provides a comprehensive framework for capturing new photos and videos via system hardware, extracting media metadata, and managing the resulting files. The library features a modular architecture that allows for custom media engine implementations to replace default image loading, file compression, and video playback logic. It offers extensive UI customization, enabling the replacement of default layout resources and theme configurations to modify

Android-Image-Cropper is an image cropping library for Android applications designed to trim and resize photos sourced from cameras or galleries. It functions as a bitmap manipulation tool capable of rotating, flipping, and scaling images to achieve specific dimensions. The library features an aspect ratio constrained cropper that limits the cropping window to fixed shapes or ratios to ensure consistent image output. It handles orientation adjustments by reading embedded metadata to automatically rotate or flip images during the cropping process. The toolkit covers a broad range of image pro

react-native-image-crop-picker is a cross-platform mobile image picker for selecting, cropping, and capturing images and videos from a device gallery or camera. It serves as a bridge to access the mobile camera and photo library to upload media files into an application. The library provides integrated tools for media cropping and compression, allowing images to be resized to specific dimensions or aspect ratios. It supports a unified workflow where images can be cropped immediately during the selection process. Capability areas include camera media capture, gallery media selection for singl

Darknet is a high-performance C-based inference engine and computer vision library designed for real-time object identification and localization. It serves as a neural network framework for training and deploying detection models using the YOLO architecture, providing a toolset for deep learning training and deployment. The project differentiates itself through a C and CUDA implementation that enables hardware acceleration for matrix multiplication and inference speed optimization. It provides a shared library interface for embedding detection capabilities into external applications and suppo

ImagePicker is a mobile image picker library that provides a user interface for selecting, rotating, and cropping images from a device gallery or camera. It includes a multi-image selection interface that allows users to pick one or multiple photos up to a specified limit. The library features a customizable image loader interface, allowing the replacement of default system image retrieval with third-party loading libraries. It also includes an image cropping tool for trimming photos into custom rectangular or circular shapes with precise dimensions and aspect ratios. The project covers medi

YOLOv9 is a real-time computer vision framework and deep learning model designed for image classification, object detection, and instance segmentation. It functions as both a vision model and a trainer, allowing for the optimization of neural network weights on custom datasets using single or multiple GPUs. The framework utilizes programmable gradient information to perform high-speed identification and location of multiple objects within images and video streams. It extends beyond bounding box detection to provide instance segmentation and panoptic segmentation, which labels every pixel in a

YPImagePicker is a set of integrated iOS components for capturing new media, browsing device galleries, and performing visual edits or filtering. It provides a built-in camera interface for photos and videos, a media selection library for picking multiple assets from the device gallery, and tools for editing media. The library includes a media editor for cropping images and trimming video durations, as well as a filter tool for modifying the visual appearance of images before they are finalized. The project covers media capture, gallery selection, and image processing. These capabilities inc