Research & Education

Find3D is an open-world 3D part segmentation model designed to identify and segment specific components within 3D objects. Users can upload their own .pcd files or select from provided samples to analyze point cloud data. The tool allows for precise part queries, enabling the segmentation of complex 3D objects into their constituent parts. This capability is particularly useful for applications requiring detailed structural analysis, object recognition, and component isolation within 3D environments. Developed as a Hugging Face Space, Find3D offers an accessible platform for researchers, developers, and enthusiasts working with 3D data and AI applications.

David-Silver-Reinforcement-learning is an open-source repository offering comprehensive notes and practical implementations for David Silver's renowned Reinforcement Learning course. It covers a wide range of topics from Week 1 (Introduction to RL) to Week 10 (Case Study: RL in Classic Games), with each week's content including slides and video links. The repository features algorithm implementations using Keras (with TensorFlow backend) and OpenAI's Gym framework, making it a valuable resource for students and researchers. It supports Python, TensorFlow, Keras, Gym, and Numpy, and encourages community contributions for expanding implementations to other frameworks like PyTorch or Caffe.

LearnClash is a competitive learning app designed to help users master any subject through engaging 1v1 quiz duels. The platform leverages AI to generate fresh, accurate questions on an infinite range of topics, from quantum physics to pop culture. Users can challenge friends or get matched with opponents of a similar skill level, with an ELO ranking system tracking progress across 8 tiers. A built-in 3-stage spaced repetition system ensures that learned information sticks, scheduling reviews at optimal intervals based on user accuracy. LearnClash is free to play, offers a premium tier for advanced features, and is available on iOS and Android.

OccNet-Course offers the first comprehensive course in China on Occupancy Network algorithms, covering everything from BEV (Bird's Eye View) to Occupancy Network principles and engineering practices, including edge-side deployment. This open-source course is designed for autonomous driving enthusiasts and professionals, providing in-depth knowledge on surrounding semantic occupancy perception. It includes detailed documentation, PowerPoint presentations, and source code, making it a valuable resource for both theoretical understanding and practical application. The curriculum covers various aspects such as BEV perception, different Occupancy Network approaches (pure vision, point cloud, multi-modal fusion), important datasets, benchmarks, and deployment strategies for NVIDIA and Horizon J5 chips. The course also features practical coding exercises and a final project to solidify learning.

Open-DiffusionGS is an open-source project that implements a novel approach to single-stage image-to-3D generation and reconstruction by integrating Gaussian Splatting directly into a diffusion denoiser. This method allows for fast and scalable creation of 3D objects, including mesh exportation, and efficient scene reconstruction without the need for depth estimators. The tool is capable of generating 3D outputs in approximately 6 seconds, significantly faster than some state-of-the-art methods. It supports both object-centric image-to-3D generation and scene-level reconstruction, with evaluation capabilities for the latter using datasets like RealEstate10K. The project provides comprehensive scripts for environment setup, quick demonstrations, data preparation for both scene and object-level datasets (including G-Objaverse), evaluation, and multi-stage training of custom models.

Pix2Text (P2T) is a free and open-source Python3 tool designed to convert visual content from images into Markdown format. It serves as an alternative to tools like Mathpix, offering core functionalities such as recognizing layouts, tables, images, text, and mathematical formulas. P2T can also convert entire PDF files, including scanned images, into Markdown. The tool integrates various models for layout analysis, table recognition, and mathematical formula detection and recognition. It supports over 80 languages for text recognition, utilizing CnOCR for English and Simplified Chinese, and EasyOCR for other languages. An online web service and demo are also available for users not familiar with Python.

Fluent was an AI-powered language learning tool designed to help users improve their language skills through interactive conversations. It aimed to facilitate passive vocabulary learning and apply comprehensible input and output by simulating real-life interactions. The tool was intended to build confidence in learners, catering to both beginners and advanced speakers. However, the project has been officially closed by its creator, who has moved on to new endeavors. Users can no longer access or utilize Fluent for language learning.

LongVU is an AI tool hosted on Hugging Face Spaces that enables users to interact with visual content by uploading videos or images and posing questions or comments. The application then processes the visual input and generates detailed text responses, providing insights and information derived from the content. This functionality makes LongVU a valuable resource for researchers and developers focused on video analysis, image understanding, and general visual content interpretation. It leverages advanced AI models to bridge the gap between visual data and textual explanations, facilitating deeper engagement with multimedia.

Hunyuan3D Part is an AI tool developed by Tencent, available through Hugging Face Spaces, designed for advanced 3D model analysis. Users can upload 3D models in common formats such as GLB, PLY, or OBJ. The tool's primary function is to segment these models into their constituent parts, providing a detailed breakdown of the object's composition. Beyond simple segmentation, it generates comprehensive part compositions and offers both segmented and exploded views of the model, which can be invaluable for design, engineering, or educational purposes. The platform currently appears to be experiencing a runtime error, preventing its full functionality from being accessed.

awesome-cs-cloudnative-blockchain is an extensive open-source repository designed as a growth handbook for individuals interested in computer science, cloud-native technologies, blockchain, web3, and Golang. It offers a curated collection of learning materials, including detailed guides on Go language, Docker, Kubernetes, and various CS fundamentals like operating systems, algorithms, and data structures. The resource also delves into blockchain technology, covering Ethereum, Bitcoin, and Hyperledger Fabric, alongside cryptography and consensus algorithms. It aims to provide a structured learning path for aspiring engineers and those looking to deepen their knowledge in these rapidly evolving fields, with content ranging from beginner to advanced topics and practical project examples.

DeepReinforcementLearning is an open-source project that replicates the AlphaZero methodology for deep reinforcement learning using Python. Developed by AppliedDataSciencePartners, this tool is designed for researchers and developers interested in exploring and experimenting with advanced AI algorithms. It provides a comprehensive framework for building and training reinforcement learning models, specifically focusing on the AlphaZero approach. The repository includes code for game environments, Monte Carlo Tree Search (MCTS), agent implementation, and model training, making it a valuable resource for understanding and applying deep reinforcement learning concepts. The project is well-suited for those looking to delve into the intricacies of AI game playing and strategic decision-making.

WolframAlpha is a powerful computational knowledge engine that provides expert-level answers and dynamic insights across a vast array of subjects. Utilizing Wolfram's breakthrough algorithms, extensive knowledgebase, and advanced AI technology, it can compute solutions for mathematics, science, technology, society, culture, and everyday life. Users can input natural language queries or mathematical expressions to receive detailed, step-by-step solutions, plots, and curated data. It's relied upon by millions of students and professionals for its ability to make the world's knowledge computable, offering a unique blend of natural language understanding, dynamic algorithmic computation, and visual representation of data.

Robotics-coursework is a GitHub repository maintained by mithi, offering a curated collection of online learning resources for robotics. It serves as a valuable directory for individuals looking to delve into robotics, whether through formal courses, textbooks, or practical projects. The repository categorizes resources into series of courses, single courses, and hands-on projects, making it easy for users to navigate based on their learning preferences. It includes links to platforms like MIT Open Courseware, Coursera, EdX, and Udemy, alongside specific university courses and practical guides for building robots with Arduino or Raspberry Pi. The repository also features sections on useful concepts, tools, and related lists, making it a comprehensive hub for robotics education.

ROS-Academy-for-Beginners is an open-source collection of code examples specifically designed for the 'Robot Operating System Introduction' course on Chinese University MOOC. This repository offers a comprehensive set of ROS packages, including robot simulation programs, various communication examples (topic, service, action, param), and demonstrations of advanced functionalities like navigation and Simultaneous Localization and Mapping (SLAM). It supports both C++ and Python implementations for many examples, making it versatile for different programming preferences. The project is actively maintained and updated, providing a valuable resource for students and developers looking to learn and implement ROS concepts. It also includes instructions for downloading, compiling, and running the examples, with specific recommendations for the operating environment.

HelloTalk is a comprehensive language learning platform designed to connect users with native speakers globally for free language exchange. With over 70 million registered users across 200+ countries, it supports learning and practicing 260+ languages through various interactive features. Users can engage in text, voice, and video chats, join live voice rooms for group practice, and participate in interactive live streams with certified teachers. The platform includes built-in AI-powered translation, grammar correction, and transliteration tools to ensure smooth conversations. Additionally, a global community feed allows users to share their language journey and receive corrections, fostering an immersive learning environment.

Jello is an innovative platform designed for creating personalized games, offering a unique blend of classic gameplay with user-generated content. Users can easily customize popular games such as Whack-A-Mole and Memory by integrating their own photos and sounds, making each game a truly personal experience. The platform emphasizes ease of use, allowing for unlimited game creation and customization without requiring any coding knowledge. Games can be shared instantly via unique links, and players do not need to download any applications or sign up to play, ensuring a seamless and accessible gaming experience. This makes Jello an ideal tool for individuals looking to create engaging, custom games for personal enjoyment, events, or educational purposes.

COBRApy is a powerful open-source Python package designed for constraint-based modeling of metabolic networks. It is widely used for genome-scale modeling in both prokaryotes and eukaryotes, offering robust infrastructure for creating and managing metabolic models. Researchers can access popular solvers and analyze models using methods such as flux balance analysis (FBA), flux variability analysis (FVA), parsimonious FBA (pFBA), and minimization of metabolic adjustment (MOMA). The tool also facilitates inspecting models to draw conclusions on gene essentiality and testing the consequences of knock-outs. COBRApy aims to be a foundational tool for developers building new COBRA-related Python packages for visualization, strain-design, and data-driven analysis, promoting re-use of classes and design principles for easier implementation and broader accessibility.

hamuleite is an open-source repository offering a comprehensive knowledge base of academic papers from prestigious institutions like National Taiwan University, National University of Singapore, Waseda University, University of Tokyo, Academia Sinica (Taiwan), and key Chinese universities and research organizations. The collection spans various disciplines including social sciences, economics, mathematics, game theory, philosophy, and systems engineering. It also includes research reports from academic and university sectors, along with summaries of academic forums in mathematics and related interdisciplinary fields. The repository is primarily intended for overseas Chinese and social science researchers, providing valuable resources for in-depth study and analysis.

Mental Math Practice Trainer is a free online platform designed to help students from class 1 to class 5 master mental math. It offers structured, timed practice for addition, subtraction, multiplication, and division, with instant feedback to build fluency and confidence. The tool focuses on accuracy before speed, allowing users to choose operations and digit levels appropriate for their grade. It features practice plans like 'Daily 10' for beginners, 'Speed 20' for intermediate learners, and 'Fluency 50' for advanced students, making it suitable for daily drills and building mental computation skills without paper or calculators. The platform also provides tips and tricks for faster mental math by operation.

china-dictatorship is an open-source GitHub repository dedicated to compiling anti-Chinese government propaganda. It serves as a comprehensive resource, featuring a mega-FAQ section that addresses common questions, a news compilation, and even recommendations for restaurants and music. The repository aims to provide information and perspectives critical of the Chinese government. It explicitly warns users in China with real names on their accounts against starring the repo to avoid police attention, highlighting the sensitive nature of its content. The project covers a wide range of topics, including censorship, human rights issues, political events, and critical analyses of key figures and policies within the Chinese Communist Party.

Biology AI: Homework Helper is a mobile application developed by Artmvstd Mobile Studio, designed to simplify complex biology topics and provide instant assistance with homework and exam preparation. The app aims to help users learn faster by generating answers to biology questions, summarizing notes, and creating study tools like quizzes and flashcards. It acts as a personalized AI tutor, offering comprehensive support for various biology learning needs. While the specific features are not detailed on the Artmvstd website, their portfolio focuses on beautifully designed mobile applications for iOS and Android, suggesting a user-friendly interface for this educational tool. The app is likely available on both Apple App Store and Google Play Store, as indicated by Artmvstd's developer links.

Noiz offers a free AI PDF summarizer that allows users to quickly generate summaries from any PDF document, regardless of size or length. The tool provides flexibility in summary output, enabling users to select their desired length (short, medium, or long) and format (bullet points, Q&A, or essay). Summaries can be downloaded as TXT or Markdown files, or simply copied to the clipboard. Noiz emphasizes its commitment to being completely free, with no hidden costs, trial periods, or feature paywalls. It supports large research papers and technical documents, processing most files within seconds, and ensures data privacy by not storing user files or summaries.

This open-source book, "Mathematical Foundations of Reinforcement Learning," offers a mathematically rigorous yet accessible introduction to the core concepts, problems, and algorithms in reinforcement learning. Designed for senior undergraduate students, graduate students, researchers, and practitioners, it requires no prior reinforcement learning background but assumes knowledge of probability theory and linear algebra. The book carefully controls mathematical depth, providing illustrative examples based on a grid world task to clarify complex ideas. It is coherently organized, building each chapter on the preceding one, and is complemented by lecture slides and a highly-viewed video series available in both Chinese and English.

Practicing-Federated-Learning is an open-source GitHub repository offering practical code examples for federated learning. It serves as a companion to the book "Practicing Federated Learning," providing hands-on implementations for various chapters. The repository covers topics such as horizontal and vertical federated learning with Python and FATE, personalized recommendations, computer vision, and advanced concepts like attack and defense mechanisms, differential privacy, and homomorphic encryption. It aims to help users understand and apply federated learning to address data silos and user privacy concerns, making it a valuable resource for both academics and industry professionals.