Data & Analytics

GigaSpeech is a comprehensive, open-source dataset specifically designed for advancing speech recognition research and development. It features over 10,000 hours of high-quality human-transcribed audio, alongside an additional 33,000+ hours suitable for unsupervised or semi-supervised learning. The dataset encompasses diverse acoustic conditions and domains, including audiobooks, podcasts, and YouTube content, with various ages and accents. It provides pre-processed versions via HuggingFace and includes detailed metadata in a version-controlled JSON file, allowing users to extract relevant information for tasks like speech recognition. GigaSpeech also offers data preparation scripts for popular toolkits like Kaldi, Espnet, and Icefall, making it easier for researchers to integrate and utilize the dataset.

Roboflow-python is an open-source Python package designed to streamline the development of computer vision applications. It provides a comprehensive set of tools for managing datasets, training models, and deploying them efficiently. The package supports a wide range of computer vision tasks, making it a versatile choice for developers working on object detection, image classification, and other related projects. Its open-source nature fosters community collaboration and allows for flexible integration into existing workflows, providing a robust foundation for building and experimenting with AI-powered vision systems.

UniDetector is an open-source computer vision tool designed for universal object detection, providing the code release for the CVPR 2023 paper "Detecting Everything in the Open World: Towards Universal Object Detection." Built upon mmdetection v2.18.0 and requiring CLIP, this tool facilitates both single-dataset and multi-dataset training, as well as open-world inference. It supports end-to-end and decoupled training/inference workflows, including probability calibration. UniDetector is ideal for researchers and developers working on advanced object detection tasks, offering robust capabilities for preparing datasets, language CLIP embeddings, and pre-trained RegionCLIP parameters.

Awesome-3D-Object-Detection is a comprehensive curated list of resources dedicated to deep learning for 3D Object Detection, with a strong emphasis on lidar-based methodologies. This GitHub repository serves as an invaluable hub for researchers and engineers, providing direct links to relevant academic papers, associated code implementations, and essential datasets like KITTI, nuScenes, Lyft, and Waymo Open Dataset. It also highlights top conferences and workshops in the field, offering a structured overview of the latest developments and trends. The resource includes surveys, books, videos, and course materials, making it a one-stop reference for anyone looking to delve into or stay current with 3D object detection.

awesome-data-annotation is a comprehensive, curated list of tools specifically designed for data annotation and management. This open-source resource categorizes tools by data type, including image/video, text, and multimodal/pointcloud, and further distinguishes between open-source and commercial options. It serves as an invaluable guide for anyone involved in preparing data for AI models, from data scientists to machine learning engineers. The list highlights popular choices like CVAT for computer vision tasks, offering detailed descriptions of each tool's capabilities, such as segmentation, geometric shapes, keypoints, and NER. It also includes UI components for integration into other applications, making it a versatile reference for developers and researchers alike.

Awesome-RGBT-Fusion is a comprehensive, open-source collection dedicated to deep learning-based RGB-T fusion methods, codes, and datasets. This resource is invaluable for researchers and developers working in computer vision, particularly those interested in multispectral data. The collection covers key areas such as Multispectral Pedestrian Detection, RGB-T Aerial Object Detection, RGB-T Semantic Segmentation, RGB-T Crowd Counting, and RGB-T Fusion Tracking. It provides access to various datasets, tools, and a curated list of academic papers with links to PDFs and code repositories. The project actively encourages contributions, making it a dynamic and evolving hub for advancements in RGB-T fusion.

Octopize is a powerful Data & Analytics tool designed to liberate the potential of sensitive, rare, or siloed data by transforming it into anonymous synthetic datasets. Utilizing a unique 'avatar' anonymization method, Octopize ensures data privacy while preserving the statistical relevance of the original information. This allows for secure data sharing, AI model development, analytics, and auditing on a trusted and compliant foundation. The platform offers flexible deployment options, including SaaS or on-premise, and guarantees 100% data sovereignty. Octopize is recognized by Gartner for its significant impact, providing a 'white box' approach that is auditable and transparent, helping organizations achieve measurable ROI by turning data into a strategic financial asset.

YData Fabric is a comprehensive platform designed to empower data scientists by improving data quality and accelerating AI model development. It offers robust features such as automated data profiling for quick exploratory data analysis, an interactive data catalog to track changes and drifts, and advanced synthetic data generation to protect sensitive information and augment datasets. The platform also provides scalable data preparation pipelines for cleaning, transforming, and orchestrating data flows, significantly reducing time-to-market for AI solutions. YData is trusted by a large community of data scientists and is recognized for its accuracy, scalability, and enterprise readiness in synthetic data.

Dog Breeds is an AI-powered tool designed to identify dog breeds from uploaded photos. Utilizing sophisticated AI algorithms, it analyzes specific breed-defining features such such as ear shape, muzzle length, fur pattern, and body size. These characteristics are then compared against a large database of over 100 known dog breeds to accurately determine your dog's breed. The service is completely free, prioritizing user privacy by not permanently storing uploaded photos. For optimal accuracy, users are advised to upload well-lit, clear photos of a single dog, with its face directly facing the camera and minimal obstructions.

Surge AI is a platform dedicated to advancing Artificial General Intelligence (AGI) by integrating the richness of human intelligence. The tool emphasizes that data, much like life experiences for humans, transforms AI into a more capable and intelligent entity. Its mission is to cultivate AGI that is curious, witty, imaginative, and brilliant, moving beyond AI optimized solely for clicks and hype. Surge AI aims to enable the development of AGI that can solve complex problems, imagine new philosophies, and drive significant advancements, such as curing cancer or unlocking scientific secrets. The platform positions itself as a crucial component in sculpting humanity's children through science and art, inviting users to contribute to this vision.

Golden Dataset, operating under ExpiredDomains.com, is a platform dedicated to the sale of premium expired .gold domains. It offers a vast selection of domains, updated daily, across numerous TLDs. The platform provides exclusive data metrics, such as estimated auction price, BrandRank, and SEO Price, alongside data from MOZ and Majestic, to help users assess domain value. While it doesn't register domains directly, it connects users to trusted registrars like GoDaddy for purchase. The tool is designed for SEOs, marketers, and investors looking for domains with authority, existing traffic, or strong brand potential, offering quick filtering and clean results.

Deep-active-learning is an open-source Python library designed for implementing and experimenting with various active learning algorithms. It provides a collection of methods such as Random Sampling, Least Confidence, Margin Sampling, Entropy Sampling, Uncertainty Sampling with Dropout Estimation, Bayesian Active Learning Disagreement, Cluster-Based Selection, and Adversarial Margin. This library is particularly useful for researchers and developers in the field of machine learning who aim to reduce the amount of labeled data required for training models while maintaining or improving performance. The repository includes prerequisites and a demo script for easy setup and experimentation, making it a practical tool for exploring active learning strategies.

Praxi is an AI-enabled data discovery and management platform designed to transform how organizations handle critical information. Many businesses still rely on informal methods like sticky notes or scattered messages, leading to inefficiencies and significant risks such as data loss, misinterpretation, and security vulnerabilities. Praxi's platform addresses these challenges by using advanced algorithms to discover hidden data sources across various formats and locations. It then maps out a comprehensive data landscape and structures the discovered data into a secure, organized system. This process supports robust data governance, ensures continuous compliance monitoring, and provides AI-ready data operations, making it particularly valuable for regulated industries like insurance and financial services.

Gender Age Detector is an AI-powered tool available as a Hugging Face Space that provides human gender and age detection from images. Users can upload a picture or paste an image link, and the application will process the input to identify individuals. For each detected person, the tool draws a bounding box and provides information on whether children, females, or males are present in the photo. This makes it a useful resource for tasks requiring basic demographic analysis from visual data. The tool is developed by Genius Society and is accessible via a web interface, making it easy to use for anyone with an internet connection.

MyVision is a free, online image annotation tool specifically designed for generating computer vision-based machine learning training data. It prioritizes user experience with features aimed at accelerating the labeling process and efficiently managing large datasets. Users can draw bounding boxes and polygons to accurately label objects within images. A key differentiator is its ability to leverage the popular 'COCO-SSD' model for automatic object annotation, operating locally in the browser to ensure data privacy. MyVision also supports importing existing annotation projects and converting datasets between various formats, making it a versatile solution for data scientists and developers working with computer vision models.

EntitySeg is an open-source toolbox designed for advanced image segmentation tasks, focusing on open-world and high-quality segmentation. It consolidates several cutting-edge algorithms developed by the qqlu group, including Open-World Entity Segmentation (TPAMI2022), High Quality Segmentation for Ultra High-resolution Images (CVPR2022), CA-SSL: Class-Agnostic Semi-Supervised Learning (ECCV2022), and High-Quality Entity Segmentation (ICCV2023 Oral). The toolbox is built using Python and PyTorch, making it accessible for researchers and developers in the computer vision domain. It aims to provide a unified platform for various image segmentation challenges, with future plans to merge all projects for enhanced interoperability and support.

golden-horse is a specialized Named Entity Recognition (NER) tool designed for Chinese social media, specifically Weibo. It offers a comprehensive dataset of 1,890 messages sampled from Weibo between November 2013 and December 2014, annotated according to DEFT ERE guidelines for both name and nominal mentions. The repository also includes a neural-based NER tool, golden-horse, which implements methods from EMNLP 2015 and ACL 2016 papers. It provides updated data with fixed inconsistencies and supplementary material for model comparison. The tool is implemented in Theano and requires Python modules like Theano and jieba for Chinese word segmentation.

Grounding Large Multimodal Model (GLaMM) is an end-to-end trained LMM designed for visual grounding, capable of processing both image and region inputs. It introduces the novel task of Grounded Conversation Generation (GCG), combining phrase grounding, referring expression segmentation, and vision-language conversations. GLaMM offers versatile interaction with visual inputs at multiple granularity levels, providing detailed region understanding and pixel-level groundings. The project also includes the GranD dataset, a large-scale, densely annotated dataset with 7.5 million unique concepts grounded in 810 million regions, each with a segmentation mask, and an automated annotation pipeline. It is open-source and available on GitHub.

Handtracking is an open-source GitHub repository that details the process and provides scripts for training a real-time hand detector using Neural Networks (SSD) on TensorFlow. It leverages the TensorFlow Object Detection API and focuses on detecting hands, particularly from an egocentric viewpoint. The project emphasizes the importance of dataset preparation, using the Egohands Dataset, and demonstrates transfer learning with models like ssd_mobilenet_v1_coco. It includes code for data cleaning, conversion to TFRecord format, and training, along with pre-trained models. The repository also highlights real-time detection capabilities on various devices and provides updates for browser-based hand tracking via Handtrack.js and Android integration using TFLite models.

CordelAI specializes in driving AI innovation by connecting top domain experts with AI development teams. Their core services include AI data labeling and curation, AI development and infrastructure annotation, and the integration of human-in-the-loop processes. They work across diverse fields such as data science, engineering, linguistics, healthcare, finance, and law to ensure AI models are trained with deep, real-world expertise. By sourcing talent across various industries, CordelAI helps build smarter, more accurate AI systems through expert annotation, content evaluation, and infrastructure support, aiming to shape a future grounded in precision, ethics, and human insight.

FineWeb-c - Annotation is an AI tool designed to streamline the process of data annotation and dataset creation. Leveraging Argilla, an open-source platform, it provides an efficient environment for labeling and annotating data. Users can provide their own datasets and utilize the tool to enhance data quality, making it suitable for training various AI models. The platform is hosted on Hugging Face Spaces, offering a convenient and accessible solution for data preparation. Its focus on efficient labeling and annotation makes it a valuable resource for anyone involved in developing and refining AI applications.

Machine-Learning-with-R-datasets is an open-source GitHub repository offering a collection of formatted datasets specifically curated for use with the book "Machine Learning with R" by Brett Lantz. This resource addresses the common issue of inaccessible datasets by providing cleaned and recoded versions that align with the book's examples. Users can easily download these public domain datasets, which include files like `challenger.csv`, `credit.csv`, and `sms_spam.csv`, directly from the repository. It serves as a practical companion for students and practitioners looking to replicate or practice machine learning techniques described in the book, ensuring they have the necessary data in the correct format without needing to purchase the book or create an account.

iCit Technologies offers comprehensive computer vision solutions designed for live monitoring, object detection, and advanced video analytics. Leveraging state-of-the-art AI technology, the platform enables users to effectively monitor their environments and proactively identify potential threats. Its capabilities extend to real-time alerts and tracking, making it a valuable asset for enhancing security systems. The technology focuses on providing robust solutions for various applications, ensuring that users have the tools to maintain situational awareness and respond swiftly to critical events. iCit Technologies aims to deliver accessible and efficient computer vision technology for diverse monitoring and analytical needs.

Object Detection Safari is a free, web-based tool designed for exploring object detection through an interactive interface. Users can search for specific objects within images by providing text prompts, or upload their own queries to find relevant images and objects. The tool delivers labeled results, offering options to refine searches for more precise outcomes. It serves as an excellent resource for individuals interested in learning about object detection, providing a hands-on experience for educational and fun exploration. Developed by MyScale, it operates as a Hugging Face Space, making it accessible for anyone to experiment with AI-powered image analysis.