Research & Education

This GitHub repository offers a Chinese translation of the second edition of the influential book "Reinforcement Learning: An Introduction." The project was initiated to provide a valuable resource for Chinese-speaking individuals interested in studying and discussing reinforcement learning concepts. While the project is now in maintenance mode due to the official Chinese translation being published, it still serves as a community-driven effort to make this complex topic more accessible. The repository includes translated chapters and aims to facilitate a deeper understanding of reinforcement learning algorithms and theories.

Awesome-holistic-3d is a valuable open-source resource for researchers and academics focused on holistic 3D reconstruction in computer vision. This GitHub repository compiles a comprehensive list of relevant papers, datasets, and code, categorized by scene-level and object-level reconstruction. It includes references to tutorials, workshops, and a wide array of research papers spanning from 2009 to 2020. The resource details various datasets with information on the number of scenes, rooms, frames, and annotated structures, making it an essential reference for anyone working on or studying 3D reconstruction techniques.

SelfExSR is a research code implementation for single image super-resolution, based on the paper "Single Image Super-Resolution from Transformed Self-Exemplars" (CVPR 2015). This algorithm stands out by achieving state-of-the-art performance in image super-resolution without requiring any external training dataset, complex feature extraction, or complicated learning algorithms. It operates by learning from transformed self-exemplars within the image itself. The repository provides the MATLAB source code, testing images for various datasets (Set5, Set14, Urban 100, BSD 100, Sun-Hays 80), and precomputed results for comparison with other state-of-the-art methods. While designed as educational code and not optimized for speed, users can adjust iteration numbers for a trade-off between speed and visual quality.

SEAM (Self-supervised Equivariant Attention Mechanism) is an open-source implementation designed for weakly supervised semantic segmentation. This tool addresses the challenge of generating accurate object masks from image-level supervision, a common limitation in advanced class activation map (CAM) solutions. SEAM introduces a self-supervised approach by enforcing consistency regularization on predicted CAMs across various transformed images, effectively narrowing the gap between full and weak supervisions. Additionally, it incorporates a pixel correlation module (PCM) to refine predictions by leveraging context appearance information and similar neighbors. Extensive experiments on the PASCAL VOC 2012 dataset demonstrate SEAM's superior performance compared to state-of-the-art methods using the same level of supervision, making it a valuable resource for AI researchers and computer vision engineers.

Uktena is an AI assistant platform that is currently experiencing service unavailability. The website indicates that the service is not accessible at this time. While the previous description suggested it was designed for various industries to improve practical knowledge sharing through digitalization and visualization, and aimed to safeguard industry expertise with isolated and secure AI environments, these features cannot be confirmed due to the current service status.

Attendance-Management-system-using-face-recognition is an open-source project built with Python and OpenCV, designed to automate attendance tracking through facial recognition. Users can register new students by taking multiple images, which are then used to train the system's facial recognition model. Once trained, the system can automatically mark attendance for registered individuals by detecting their faces. It generates CSV files for attendance records, organized by subject, and allows users to view attendance data in a tabular format. This system requires users to set up their environment and adjust file paths, making it a technical solution for automated attendance.

Zero Shot Text Classification is an AI tool hosted on Hugging Face Spaces by datasciencedojo, designed for classifying text into predefined categories without requiring specific training data for those categories. Users can easily input a piece of text and provide a list of candidate labels or categories. The tool then processes the input and returns a score for each category, indicating how well the text fits into that particular classification. This makes it a highly flexible and efficient solution for quick text categorization tasks, eliminating the need for extensive dataset preparation and model training.

reinforcement_learning_course_materials offers comprehensive lecture notes, tutorial tasks including solutions, and online videos for a reinforcement learning course. Originally hosted at Paderborn University and now transferred to the University of Siegen, this open-source material is licensed under a Creative Commons Attribution 4.0 International Public License. It is designed for both self-learning students and lecturers looking to set up their own courses. The content covers a wide range of topics from introduction to reinforcement learning, Markov decision processes, dynamic programming, Monte Carlo methods, and various policy gradient methods, all with accompanying video lectures and practical exercises based on Python 3.12.

earth2studio is an open-source, Python-based deep-learning framework developed by NVIDIA, designed to accelerate the exploration, building, and deployment of AI-driven weather and climate workflows. It offers a unified API for various AI frameworks, model architectures, and data sources, promoting composability and rapid development of complex pipelines. Key features include a comprehensive model zoo with state-of-the-art prognostic and diagnostic AI weather/climate models, standardized data source access (GFS, ERA5, IFS), and flexible I/O backends (Zarr, NetCDF4). The framework also provides perturbation methods for ensemble forecasting and statistical operations for in-pipeline evaluation, making it a powerful toolkit for researchers and developers in climate science.

Gemini Playground is a Hugging Face Space developed by Roboflow, offering an interactive platform to engage with Gemini Pro models. Users can upload images and type messages to receive detailed responses, making it ideal for experimenting with multimodal AI capabilities. The tool provides options to adjust the response style and length, allowing for customized interactions. Built with Gradio, it offers a user-friendly interface for AI enthusiasts, developers, and researchers to test and prototype AI applications, exploring the potential of Gemini Pro in various scenarios.

pytorch-pose is an open-source PyTorch toolkit designed for 2D single human pose estimation. It offers a comprehensive pipeline for training, inference, and evaluation, making it a valuable resource for researchers and developers in computer vision. The toolkit includes a robust dataloader with various data augmentation options, compatible with popular human pose databases such as MPII, LSP, and FLIC. Key features include multi-thread data loading, multi-GPU training support, a logger for tracking progress, and visualization of training and testing results. It is compatible with PyTorch 0.4.1/1.0 and provides detailed instructions for installation, data preparation, and usage, including testing with pre-trained models and evaluating PCKh@0.5 scores.

CVPR-2019-Paper-Statistics is an open-source project offering detailed statistics and visualizations for papers accepted at the CVPR 2019 conference. Inspired by ICLR2019-OpenReviewData, this tool analyzes the acceptance rate trends from 2015 to 2019, highlighting the significant increase in paper submissions and the corresponding decrease in acceptance rates. It also provides insights into the most frequent keywords in accepted papers, such as 'Image', 'detection', '3d', 'object', 'video', 'segmentation', 'adversarial', 'recognition', and 'visual'. The project includes Jupyter Notebook code for analysis and visualization, supporting both CSV and website data formats, and requires Python 3.5 with libraries like selenium, wordcloud, and matplotlib.

MARS (Modular and Realistic Simulator for Autonomous Driving) is an open-source project designed to provide an instance-aware, modular, and realistic simulation environment for autonomous driving research. It allows users to train and test autonomous vehicle algorithms using various datasets like KITTI and vKITTI2. The simulator supports reconstruction and novel view synthesis tasks, offering pre-trained models and the flexibility to train from scratch with custom data. Its modular framework enables combining different architectures for various nodes, such as using Nerfacto for background models. MARS is built upon Nerfstudio and requires an NVIDIA GPU with CUDA for installation and operation.

Blarma is a mobile application designed to facilitate vocabulary acquisition and language learning through a scientifically proven, four-step method. It leverages visual and audio training, pairing new words with images to trigger dual coding for intuitive understanding. The app places words into AI-personalized sentences and stories, promoting contextual learning similar to real-life language acquisition. Blarma incorporates a smart spaced repetition algorithm to manage the 'forgetting curve,' ensuring long-term retention of vocabulary. Users can engage in unlimited practice drills for pronunciation, writing, and listening, transforming passive knowledge into active language skills. It supports 14 languages and is available on iOS and Android.

Embodied_AI_Paper_List is an open-source repository maintained by HCPLab at SYSU and Pengcheng Laboratory, offering a comprehensive collection of papers and resources focused on Embodied AI. This resource is designed to serve as a foundational reference for researchers and practitioners, bridging the gap between cyberspace and the physical world through intelligent systems. The repository covers key areas such as embodied perception, interaction, agent development, and sim-to-real adaptation, including state-of-the-art methods, essential paradigms, and comprehensive datasets. It also explores the role of Multi-modal Large Models (MLMs) and World Models (WMs) in facilitating interactions for embodied agents, highlighting their significance in both digital and physical environments. The list is regularly updated with the latest advancements and includes a survey paper accepted by IEEE/ASME Transactions on Mechatronics.

FlashWorld Demo Spark provides a user-friendly interface for interacting with the FlashWorld environment, enabling the creation of dynamic 3D scenes. Users can define camera paths and enrich their scenes with various prompts, including images or detailed text descriptions. The tool allows for comprehensive configuration of settings and the recording of camera movements, streamlining the scene generation process. Designed for ease of use, it facilitates the rapid creation of immersive 3D content, making advanced 3D scene generation accessible to a broader audience.

Florence 2 is an AI tool developed by HuggingFaceM4 that enables users to interact with images by asking questions. Users can upload an image and provide a text prompt to query the image, and the application will generate an answer based on the visual content and the contextual information given. This tool is designed for image-based question answering, allowing for a deeper understanding and extraction of information from visual data. It is offered as a free-to-use application, licensed under Apache-2.0, making it accessible for various applications including research and educational purposes.

Low-effort papers functions as a specialized Google Scholar search aggregator, designed to pinpoint and categorize academic papers that might be considered 'low-effort.' While the exact criteria for 'low-effort' are not detailed, the tool aims to assist researchers in rapidly conducting literature reviews and gaining insights into publication patterns within specific research domains. This can be particularly useful for those looking to quickly grasp the landscape of a topic without delving into highly complex or extensive studies. The platform leverages Google Scholar's vast database to streamline the discovery process, potentially saving time for academics, students, and professionals who need to survey existing research efficiently.

introRL offers a comprehensive introduction to reinforcement learning, featuring a series of 10 lectures with accompanying slides. The course content is presented in English slides, while the lectures are delivered in Mandarin by Bolei Zhou, making it accessible to a broad audience interested in the subject. It covers fundamental topics such as Markov Decision Processes, model-free prediction and control, value function approximation, and policy optimization. Additionally, it delves into advanced concepts like model-based RL, imitation learning, and distributed systems for RL, concluding with a summary and a bonus lecture on DeepMind's AlphaStar. This resource is ideal for individuals seeking to understand the core principles and advanced applications of reinforcement learning for personal educational purposes.

Fairly Multilingual ModernBERT Token Alignment is an AI tool designed to align words between two sentences across multiple languages. Users can input sentences in English, French, Dutch, or German, and the application will identify and display corresponding words between them. This functionality is particularly useful for linguists, translators, and researchers working with multilingual texts, enabling detailed comparative analysis of sentence structures and word usage. Built with Streamlit and available as a Hugging Face Space, it offers an accessible platform for facilitating multilingual analysis and linguistic research.

The FAIR Chemistry Leaderboard is a Hugging Face Space developed by Facebook for chemistry research. It enables researchers to upload their model prediction files, such as NPZ or JSON, along with essential information about their model and training dataset. The platform then evaluates these predictions against established reference data, providing a standardized way to track and compare model performance. This tool is designed to foster progress in chemistry-related tasks by offering a transparent and collaborative environment for benchmarking AI models in the field. It is built with Gradio and requires Hugging Face authentication for access.

Fast Sd3.5 Large is an AI application hosted on Hugging Face Spaces, designed to execute Python scripts provided by the user. Users need to set the 'MY_SCRIPT_CONTENT' environment variable with their desired Python script, and the application will then run this script. This setup offers a flexible environment for developers and researchers to test and deploy custom AI models or scripts without managing the underlying infrastructure. It's particularly useful for quick experimentation and sharing Python-based AI functionalities within the Hugging Face ecosystem.

Filechat is an AI-powered tool designed to help users interact with their documents. Users can upload various documents and then engage with a chatbot to ask questions about the content. The chatbot is capable of providing precise answers, complete with direct citations from the uploaded material, ensuring accuracy and traceability. Filechat offers different subscription plans, which include credits for various features, such as API integration and secure cloud storage, catering to different user needs.

FreshFeed is an AI tool designed to function as a search engine specifically for Large Language Models (LLMs). Its primary objective is to enhance the accuracy and reliability of LLMs by supplying them with current information, thereby mitigating the issue of hallucinations. The platform is currently in its development phase, with its website indicating that it is under construction. Users are advised to check back for updates soon, as the service is not yet live or accessible.