Research & Education

The 3D Web Visualizer provides a real-time, interactive 3D representation of a Reachy Mini robot directly in a web browser. This application allows users to rotate, zoom, and pan the robot model, with its joint angles updating dynamically to reflect the robot's actual pose. It's designed for visualizing live robot data, making it suitable for remote monitoring, educational demonstrations, and development purposes. The tool offers a clear and intuitive way to observe robot movements and configurations without needing specialized software, enhancing accessibility for various users.

ACL Pubcheck is a specialized tool designed to assist researchers in ensuring their academic papers meet the specific guidelines for ACL (Association for Computational Linguistics) conferences. Hosted on Hugging Face Spaces, this platform offers a user-friendly graphical interface where users can easily upload their research papers in PDF format. After selecting the appropriate paper type, the tool processes the document to identify any non-compliance issues, providing valuable feedback to authors. This helps streamline the submission process by catching potential formatting or content errors before official submission, making it an essential resource for academics in the computational linguistics field.

Concerto is an AI tool available on Hugging Face Spaces that specializes in reconstructing 3D scenes from input video or PLY point-cloud files. The application leverages advanced depth and pose estimation techniques to generate a detailed 3D point cloud representation of the scene. A unique feature of Concerto is its application of Principal Component Analysis (PCA) to color the points within the reconstructed cloud, which helps in highlighting different aspects or features of the scene. This tool is particularly useful for researchers and developers working with 2D-3D self-supervised learning and spatial representations, offering a practical way to visualize and analyze complex spatial data. It provides a hands-on demonstration of the concepts presented in the paper "Concerto: Joint 2D-3D Self-Supervised Learning Emerges Spatial Representations."

The Chinese Open Source Heatmap is an interactive tool hosted on Hugging Face, designed to visualize the release activity of open-source models by Chinese AI labs. Users can view a heatmap that illustrates the number of models each lab has released over the past year. Beyond the pre-generated overview, the tool allows users to search for any Hugging Face organization or individual user to create a custom heatmap, providing a personalized view of their open-source contributions. This makes it a valuable resource for researchers, developers, and anyone interested in tracking the contributions and trends within the Chinese AI open-source community. The tool is straightforward to use, offering a clear and visual representation of data.

GIFT Eval, a Hugging Face Space by Salesforce, is a comprehensive tool designed for researchers and academics to evaluate and compare the performance of language models. It offers a user-friendly interface to browse results across various benchmarks, providing insights into model efficacy. Users can switch between overall performance metrics and detailed views, segmenting data by domain, frequency, term length, and variate type. This flexibility makes it an invaluable resource for understanding the nuances of language model behavior and identifying strengths and weaknesses across different contexts. Built with Gradio, GIFT Eval aims to facilitate advanced research in time series forecasting and language model analysis.

The Ginkgo AbDev benchmark is a specialized tool designed for the 2025 Antibody Developability Competition. Hosted on Hugging Face Spaces, this application offers a dynamic leaderboard where users can explore and analyze submitted models. Researchers and participants can sort, filter, and search through various models based on specific metrics or properties relevant to antibody development. The platform also provides the functionality to download the GDPa1 dataset, making it a valuable resource for those involved in antibody design and research. It serves as a central hub for competition participants to track progress and compare different approaches.

UniAD is a unified autonomous driving algorithm framework developed by OpenDriveLab, distinguished by its planning-oriented philosophy. Unlike traditional modular designs, UniAD hierarchically integrates perception, prediction, and planning tasks into a single framework. This approach has enabled UniAD to achieve state-of-the-art performance across all these tasks, particularly in motion prediction, occupancy prediction, and planning, with impressive metrics like 0.71m minADE for motion and 0.31% avg.Col for planning. The framework is open-source, available on GitHub, and has received the CVPR 2023 Best Paper Award. It supports integration with datasets like nuPlan and NAVSIM, and offers tools for CARLA and closed-loop evaluation. UniAD is designed for researchers and developers in the autonomous driving domain, providing a robust platform for advancing self-driving technology.

UDTL is an open-source repository providing the implementation details for the paper "Applications of Unsupervised Deep Transfer Learning to Intelligent Fault Diagnosis: A Survey and Comparative Study." It serves as a comprehensive library for researchers and academics interested in applying unsupervised deep transfer learning (UDTL) to intelligent fault diagnosis. The project offers baseline accuracies and a unified framework, allowing users to load their own datasets and models for new studies. It includes various loss functions for mapping-based DTL, data augmentation methods, PyTorch datasets for time and frequency domains, and models used in the project. The repository also provides utilities for the training procedure, making it a valuable resource for replicating and extending research in this field.

Video Analyst is an open-source project from Megvii Research that provides a collection of fundamental algorithms for video understanding tasks. It specifically focuses on Single Object Tracking (SOT) and Video Object Segmentation (VOS). The tool includes implementations like SiamFC++ for robust and accurate visual tracking and a State-Aware Tracker for real-time video object segmentation. It is designed for researchers and developers, offering detailed documentation for setup, model usage, training, and testing. The repository structure is well-organized, with separate modules for experiments, data handling, model building, and pipeline construction, making it a valuable resource for those working on advanced computer vision and video analysis projects.

semantic-segmentation is an open-source PyTorch library designed for state-of-the-art semantic segmentation models. It provides a flexible and customizable framework for computer vision researchers and developers. The library supports a wide array of datasets, making it suitable for various applications requiring precise pixel-level classification. Its focus on ease of use and customizability allows users to adapt models to specific needs, ensuring high accuracy for diverse computer vision projects. This tool is ideal for those looking to implement or experiment with advanced semantic segmentation techniques.

4DGaussians is a research project presented at CVPR 2024, focusing on 4D Gaussian Splatting for real-time dynamic scene rendering. This method allows for very quick convergence and achieves real-time rendering speeds, as demonstrated on D-NeRF and HyperNeRF datasets. The project provides code for environmental setup, data preparation for synthetic and real dynamic scenes (D-NeRF, HyperNeRF, DyNeRF, and multiple views), training, rendering, and evaluation. It also includes helpful scripts for exporting 3D Gaussians, visualizing weights, and merging 4D Gaussians, making it a comprehensive resource for researchers in computer vision and graphics.

awesome_lists is an open-source GitHub repository designed to be a comprehensive resource hub for Tenure-Track Assistant Professors (TTAPs) and PhD students, particularly in Computer Science. Curated by a recently graduated CS PhD and incoming TTAP, it offers valuable lists covering a wide array of topics essential for academic and professional success. These include funding and grant resources, computational resource comparisons (like GPU cost-compute trade-offs), academic social media and profile management, conference timelines, workshops and competitions, lab management advice, and general guidance from senior academics. The project aims to be a living document, continuously refined with community contributions, and is especially useful for those navigating the early stages of their academic careers.

Awesome_Prompting_Papers_in_Computer_Vision is a comprehensive, curated list of research papers focusing on prompt-based techniques within the fields of computer vision and vision-language learning. This resource is designed to help researchers and practitioners stay abreast of the rapidly evolving advancements in visual prompting. It categorizes papers into key areas such as Vision Prompt, Vision-Language Prompt, Language-Interactable Prompt, and Vision-Language Instruction Tuning. Each entry typically includes links to the paper and often to associated code, making it a valuable hub for exploring foundational models, parameter-efficient adaptation, and multimodal learning approaches.

awesome-embedded-and-iot-security is a comprehensive, curated list of resources dedicated to embedded and IoT security. This open-source project aims to assist both beginners and experts in navigating the complex landscape of securing embedded and Internet of Things devices. The list encompasses a wide array of resources, including software tools for analysis, extraction, and support, as well as various hardware tools for Bluetooth BLE, ZigBee, SDR, and RFID/NFC. Additionally, it features an extensive collection of books, research papers, and case studies to provide theoretical and practical insights. The resource also points to free training, websites, blogs, tutorials, YouTube channels, and conferences, making it a one-stop-shop for anyone looking to enhance their knowledge or conduct their own security analysis in this critical domain.

ChatGod is described as being associated with Zenith's ZOI satellite, a prototype specifically engineered for scientific experiments conducted in orbit. This satellite is equipped with a pressurized payload compartment, enabling a variety of microgravity experiments. Beyond its experimental capabilities, ChatGod also supports remote sensing and advanced communications technologies. The satellite's mission is designed for a duration of six months in orbit, focusing on data collection and scientific research. The tool's connection to such a specialized satellite suggests its application in highly technical and scientific domains, likely catering to researchers and institutions involved in space science and advanced technological development.

Maml is an open-source code repository for Model-Agnostic Meta-Learning (MAML), a technique designed for the fast adaptation of deep networks. Developed by cbfinn, this repository provides the foundational code accompanying the paper "Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks" (Finn et al., ICML 2017). It specifically includes implementations for few-shot supervised learning domain experiments, covering tasks such as sinusoid regression, Omniglot classification, and MiniImagenet classification. The project is built using Python 2.* or 3.* and TensorFlow v1.0+, making it accessible for researchers and developers working in meta-learning and few-shot learning. Users can access data preparation instructions for Omniglot and MiniImagenet, and detailed usage instructions are available within the `main.py` file.

deep-rl-class is the official GitHub repository for the Hugging Face Deep Reinforcement Learning Course. This open-source resource offers comprehensive materials, including mdx files and Jupyter notebooks, designed to teach both the theoretical and practical aspects of Deep Reinforcement Learning. While the course is currently in a low-maintenance state, it remains an excellent educational resource. Users can access the full syllabus and course content, though some features like Unit 7 (AI vs AI) and the Leaderboard are non-functional. The repository encourages community engagement for problem-solving in hands-on exercises.

MMSkeleton is an open-source toolbox developed by OpenMMLAB, specifically designed for skeleton-based human understanding. It offers a highly extensible framework that systematically organizes code and projects, allowing for adaptation to various tasks and scaling to complex deep models. Key functionalities include 2D and 3D pose estimation, skeleton-based action recognition (like ST-GCN), and action synthesis. The toolbox also supports building custom skeleton-based datasets and creating personalized applications. It is part of the OpenMMLAB project, developed on the ST-GCN research project, and is released under the Apache 2.0 license.

evolution-strategies-starter offers a distributed implementation of the Evolution Strategies algorithm, as detailed in the paper "Evolution Strategies as a Scalable Alternative to Reinforcement Learning." This open-source project utilizes a master-worker architecture where the master broadcasts parameters to workers, and workers return results. The code is specifically designed to run on AWS EC2, making it resilient to worker termination and suitable for spot instances. It requires a Mujoco license for humanoid experiments and uses Packer for AMI building. The project is provided as-is, with no further updates expected, serving as a foundational codebase for researchers and developers in reinforcement learning.

Model Comparator Space Builder is an AI tool designed for comparing various AI models. It provides a platform for researchers and data scientists to effectively evaluate the performance of different models and benchmark their results against each other. This tool is instrumental in the model selection process, helping users make informed decisions based on comparative analysis. It supports research and development efforts by offering a structured environment for model assessment, which is crucial for advancing AI applications. The tool aims to streamline the process of understanding model strengths and weaknesses, contributing to more robust and efficient AI solutions.

Number Recognizer is an AI tool hosted on Hugging Face that specializes in recognizing digits from images of house or door plates. Users can easily upload a picture containing a house or door number, select a preferred model checkpoint, and the application will quickly process the image to read the displayed digits. The tool then returns the recognized number as plain text, along with a status indicating the recognition outcome. This application is useful for tasks requiring automated number extraction from real-world images, offering a straightforward solution for digit recognition.

Skywork-R1V is an advanced multimodal AI model series developed by Skywork AI, specializing in vision-language reasoning. The series includes both open-source versions with model weights and inference code, as well as closed-source offerings like Skywork-R1V4-Lite. These models deliver exceptional performance across vision understanding, code execution, and deep research tasks, featuring agentic capabilities. Key features include code execution for complex tasks, deep research integration with web search, multi-turn reasoning with tool usage, and streaming support for real-time responses. The models have demonstrated state-of-the-art performance on various multimodal benchmarks, particularly excelling in perception and deep research capabilities.

ObjectPoseEstimationSummary is a comprehensive GitHub repository dedicated to curating resources for object pose and viewpoint estimation. It serves as a central hub for researchers and practitioners, offering a meticulously organized collection of papers, datasets, and rendering methods relevant to the field. The repository categorizes resources into 'Objects in the wild,' 'Objects in controlled environments,' and '3D model datasets,' providing detailed annotations, statistics, and references for each entry. It also includes information on various rendering methods, such as differentiable renderers and physical simulators. The project is open-source, welcoming contributions and suggestions to further enrich its content, making it an invaluable tool for academic research and development in computer vision.

New-View-Synthesis is a comprehensive GitHub repository dedicated to collecting and organizing research papers focused on new view synthesis techniques. The repository serves as a valuable resource for researchers and academics, offering direct links to published papers (often via arXiv or PDF) and their corresponding code implementations. It is actively maintained, with daily updates to include the latest advancements and provide more detailed information about each paper. This makes it an essential tool for staying current with the rapidly evolving field of neural radiance fields and other view synthesis methodologies, facilitating research, development, and understanding of these complex topics.