Research & Education

Smart-money-concepts is a Python package designed for algorithmic trading, integrating Inner Circle Trader (ICT) concepts into Python. It provides a suite of indicators such as Fair Value Gap (FVG), Swing Highs and Lows, Break of Structure (BOS) & Change of Character (CHoCH), Order Blocks (OB), and Liquidity. The package also includes functionalities to identify previous highs and lows across different timeframes and to analyze session-specific market activity and retracements. This tool is intended for traders and investors seeking to gain deeper insights into market sentiment, trends, and potential reversals through programmatic analysis.

AiiDA (Automated Interactive Infrastructure and Database for computational science) is a powerful open-source workflow manager designed for computational science. It emphasizes robust data provenance tracking, high performance, and extensibility, allowing researchers to manage complex computational workflows efficiently. Key features include the ability to write complex, auto-documenting workflows in Python, an event-based workflow engine supporting thousands of processes per hour with full checkpointing, and automatic tracking of inputs, outputs, and metadata for full reproducibility. AiiDA also offers a flexible HPC interface compatible with various schedulers like SLURM and PBS Pro, a plugin interface for extending functionality with new simulation codes and data types, and tools for open science, enabling the export and sharing of provenance graphs.

AliceVision is an open-source photogrammetric computer vision framework designed for 3D reconstruction and camera tracking. It provides a robust software foundation with state-of-the-art computer vision algorithms that can be tested, analyzed, and reused. The project is a collaborative effort between academia and industry, ensuring cutting-edge algorithms meet the quality and robustness required for production use. It allows users to infer the geometry of a scene from a set of unordered photographs or videos, effectively reversing the 3D scene to 2D projection process. The framework is primarily used through Meshroom, which offers both a user interface and a command-line tool for launching the AliceVision pipeline and customizing workflows with Python scripting.

This GitHub repository, Assemblies-of-putative-SARS-CoV2-spike-encoding-mRNA-sequences-for-vaccines-BNT-162b2-and-mRNA-1273, serves as a vital resource for researchers studying RNA vaccines. It provides experimental sequence information for the RNA components of the Moderna (mRNA-1273) and Pfizer/BioNTech (BNT-162b2) COVID-19 vaccines. The project details the methodology used for obtaining and sequencing these RNAs from discarded vaccine remnants, including extraction, fragmentation, and sequencing protocols. It offers assembled contigs encoding full-length spike proteins, verifying the Pfizer sequence and providing a working assembly for Moderna's vaccine. This data is crucial for researchers and clinicians to identify therapeutic-derived RNA sequences in high-throughput RNA-seq studies, distinguishing them from host or infectious origins.

Astrometry.net is an open-source project designed for the automatic recognition and calibration of astronomical images. It takes an input image and returns astrometric calibration metadata, along with lists of known celestial objects within the field of view. This tool is invaluable for astronomers and researchers who work with images where celestial coordinates are unknown or untrusted, helping to organize, annotate, and make astronomical information searchable. The project is developed on GitHub and offers documentation, a web service, and Docker containers for easy deployment and use.

This open-source project, Autonomous-Driving-in-Carla-using-Deep-Reinforcement-Learning, focuses on training an autonomous driving agent using Deep Reinforcement Learning (DRL) within the CARLA urban simulation environment. It specifically employs the Proximal Policy Optimization (PPO) algorithm for learning complex decision-making tasks in a continuous state and action space. A key feature is the integration of a Variational Autoencoder (VAE) to compress high-dimensional observations into a low-dimensional latent space, potentially accelerating the agent's learning process. The project provides an end-to-end solution for autonomous driving, covering CARLA environment setup, VAE implementation, and PPO agent training. It includes pre-trained PPO agents for different CARLA towns and detailed instructions for setting up the project, installing dependencies, and running or training new agents.

Awesome-3D-Object-Detection-for-Autonomous-Driving is a GitHub repository that accompanies a comprehensive survey paper titled "3D Object Detection for Autonomous Driving: A Comprehensive Survey (IJCV 2023)". This resource is designed to help researchers and engineers stay updated on the latest advancements in 3D object detection techniques for autonomous driving systems. The repository categorizes methods into LiDAR-based, Camera-based, Multi-Modal, Temporal, and Label-Efficient 3D Object Detection, as well as their application in Driving Systems. It provides detailed overviews of various approaches within each category, including point-based, grid-based, anchor-based, and fusion techniques. The content is structured to offer a chronological overview and includes links to relevant papers, making it an essential reference for anyone working in this specialized domain.

Mapanything Gradio is a Hugging Face Space application developed by Facebook that specializes in generating detailed 3D models and depth maps from uploaded images or videos. The tool reconstructs visual data, offering comprehensive visualizations of both depth and surface normals. This functionality enables users to perform precise measurements directly on the reconstructed 3D models. Hosted on Hugging Face, it leverages advanced AI capabilities to transform 2D inputs into rich 3D representations, making it a valuable resource for tasks requiring spatial analysis and 3D reconstruction from visual media.

avod is an open-source implementation of the Aggregate View Object Detection (AVOD) network, specifically designed for 3D object detection in autonomous driving scenarios. This repository offers a Python-based solution for researchers and developers to implement and experiment with advanced 3D object detection algorithms. It leverages view aggregation techniques to enhance detection accuracy. The project includes detailed instructions for setting up the environment, installing dependencies, configuring training parameters, and running evaluations on datasets like KITTI. It also provides pre-trained models and scripts for visualizing results, making it a comprehensive resource for those working in the field of autonomous vehicle perception.

jsfeat is an open-source JavaScript Computer Vision library designed for developers to explore and implement modern computer vision algorithms using JS/HTML5. The library provides a comprehensive set of features, including custom data structures and essential image processing methods such as grayscale conversion, box blur, Gaussian blur, histogram equalization, Canny edges, and various derivative calculations. It also incorporates a Linear Algebra module for LU, Cholesky, and SVD solvers, along with Eigen Vectors and Values. For advanced applications, jsfeat offers a Multiview module with Affine2D and Homography2D motion kernels, and RANSAC/LMEDS motion estimators. Additionally, it includes feature detectors like Fast Corners, YAPE06, YAPE, and ORB, as well as Lucas-Kanade optical flow and HAAR/BBF object detectors, making it a versatile tool for computer vision development.

SegLossOdyssey is an open-source repository offering a comprehensive collection of loss functions specifically designed for medical image segmentation. This tool is invaluable for researchers and practitioners aiming to enhance the accuracy and robustness of their segmentation models, particularly in tasks involving highly imbalanced data. The collection includes implementations in PyTorch and Keras, covering a wide array of loss functions from various research papers and challenges. It highlights the effectiveness of compound loss functions for challenging segmentation tasks and provides a valuable resource for exploring and applying state-of-the-art loss functions in medical imaging.

LIBERO is an open-source benchmarking framework designed for studying knowledge transfer in multitask and lifelong robot learning problems. It provides a procedural generation pipeline capable of creating an infinite number of manipulation tasks, alongside 130 pre-defined tasks grouped into four distinct task suites: LIBERO-Spatial, LIBERO-Object, LIBERO-Goal, and LIBERO-100. These suites are structured to facilitate research into specific types of knowledge transfer, with LIBERO-100 focusing on entangled knowledge transfer for pretraining and testing lifelong learning performance. The framework also includes five research topics, three visuomotor policy network architectures, and three lifelong learning algorithms, along with sequential finetuning and multitask learning baselines. High-quality human teleoperation demonstrations are available for all task suites.

Learnable-triangulation-pytorch is an official PyTorch implementation of the paper "Learnable Triangulation of Human Pose" (ICCV 2019, oral). This open-source project focuses on 3D human pose estimation from multiple cameras, offering two novel methods: Algebraic and Volumetric learnable triangulation. These methods significantly outperform previous state-of-the-art techniques, with the Volumetric model achieving a 2.4 times reduction in error. The repository provides code for training and evaluation, supports both single and multi-GPU setups, and includes pretrained models and configurations for the Human3.6M dataset. It is designed for researchers and engineers working on advanced computer vision tasks, particularly in human pose estimation.

darknet_ros is a ROS (Robot Operating System) package designed for real-time object detection in camera images, leveraging the You Only Look Once (YOLO) system. It supports YOLO V3 on both GPU and CPU, offering significant speed advantages with CUDA-enabled GPUs. The package comes with pre-trained models capable of detecting objects from VOC and COCO datasets, and also allows users to train and deploy networks with their own custom detection objects. It provides ROS-related parameters for configuring publishers, subscribers, and actions, making it highly adaptable for robotics applications. The tool is open-source and actively maintained by leggedrobotics, providing a robust solution for integrating advanced object detection into robotic systems.

neural-combinatorial-rl-pytorch offers a PyTorch implementation of Neural Combinatorial Optimization with Reinforcement Learning, based on the research paper. This open-source tool provides a basic RL pretraining model that utilizes greedy decoding. A notable feature is its use of an exponential moving average critic instead of a traditional critic network, which has been shown to significantly improve results, particularly for the Traveling Salesperson Problem (TSP). The implementation supports a stochastic decoding policy during training and beam search for testing. It currently includes support for a sorting task and the planar symmetric Euclidean TSP, with clear guidelines for extending it to other combinatorial optimization problems by providing a dataset class and a reward function. The repository also details dependencies and provides performance results for both TSP and sorting tasks, demonstrating its generalization capabilities.

FishNet offers the implementation code for the FishNet architecture, a versatile backbone designed for image, region, and pixel-level prediction tasks. Based on a NeurIPS 2018 paper, this tool provides pre-trained models with varying parameters and FLOPs, including FishNet99, FishNet150, and FishNet201, with reported Top-1 and Top-5 accuracies. It supports training with PyTorch and includes configurations for data augmentation methods like random flip, random crop, and random PCA lighting. The project also details how to load and utilize these models, making it a valuable resource for researchers and developers working on computer vision challenges.

FaceRecognitionApp is an open-source Android application developed by Kristian Lauszus in 2016, designed to showcase face recognition capabilities. The app implements Eigenfaces and Fisherfaces algorithms for facial recognition, leveraging the FaceRecognitionLib library for its core calculations. It provides a practical example for developers interested in integrating face recognition into Android applications. The project is released under the GNU General Public License, encouraging community contributions and modifications. It requires Android Studio, the Android NDK, OpenCV Android SDK, and Eigen3 libraries for building and running, with detailed instructions provided for both basic and advanced users who wish to modify the source code.

ExtremeNet is an open-source object detection system that employs a bottom-up approach to identify objects within images. It achieves this by detecting four extreme points (top-most, left-most, bottom-most, right-most) and one center point of objects using a standard keypoint estimation network. These five keypoints are then grouped into a bounding box if they are geometrically aligned. This method transforms object detection into a purely appearance-based keypoint estimation problem, bypassing region classification or implicit feature learning. The project is built upon the CornerNet code and integrates code from Deep Extreme Cut (DEXTR) for instance segmentation, allowing it to generate coarse octagonal masks and further refine them for improved Mask AP. It provides code for training, evaluation, and demo purposes, supporting benchmark evaluation on datasets like MS COCO.

tf-image-segmentation is an open-source image segmentation framework built upon Tensorflow and the TF-Slim library. Its core purpose is to streamline the process of converting various image segmentation datasets, including general, medical, and other types, into a unified and easy-to-use .tfrecords format for training. The framework includes a robust training routine that supports on-the-fly data augmentation, such as scaling and color distortion, ensuring effective model training. It also provides functionalities for evaluating model accuracy using common metrics like Mean IOU, Mean pixel accuracy, and Pixel accuracy. The framework offers pre-trained model files and definitions for models like FCN-32s, FCN-16s, and FCN-8s, initialized with weights from Image Classification models like VGG, making it a comprehensive solution for researchers and developers working on image segmentation tasks.

Tiny Differentiable Simulator is a header-only C++ and CUDA physics library designed for reinforcement learning and robotics applications. It boasts zero dependencies, making it a lightweight and efficient solution for developers. The library implements various rigid-body dynamics algorithms, including forward and inverse dynamics, alongside contact models based on impulse-level LCP and force-based nonlinear spring-dampers. It also includes actuator models for motors, servos, and Series-Elastic Actuator (SEA) dynamics. The entire codebase is templatized, supporting automatic differentiation scalar types like CppAD, Stan Math fvar, and ceres::Jet, as well as regular float/double precision and fixed-point integer math for cross-platform deterministic computation. It can run thousands of simulations in parallel on a single RTX 2080 CUDA GPU at 50 frames per second and offers OpenGL 3+ and MeshCat visualizers.

ProtoMotions is a GPU-accelerated simulation and learning framework designed for training physically simulated digital humans and humanoid robots. It serves as a fast prototyping platform for researchers and practitioners in animation, robotics, and reinforcement learning, bridging efforts across these communities. The framework emphasizes modularity, extensibility, and scalability, allowing users to train fully physically simulated characters from large motion datasets within hours using multiple GPUs. Key capabilities include one-command retargeting of motion data to various robots, training robots to perform motor skills, and sim-to-sim testing across different physics engines like NVIDIA Newton and MuJoCo. ProtoMotions also supports sim-to-real deployment, enabling policies trained in simulation to transfer directly to real hardware like the Unitree G1 humanoid robot. It offers high-fidelity rendering in IsaacSim and integration with motion authoring tools like Kimodo for text-to-motion generation.

Face Mesh Workflow is a tool hosted on Hugging Face Spaces that allows users to upload an image, detect faces within it, and generate a 3D mesh. It offers the flexibility to adjust depth sources and customize the generated mesh using various sliders. The primary output is an OBJ file, which can then be downloaded for further use in other 3D modeling or animation software. This tool is particularly useful for those working with facial recognition, 3D modeling, or anyone needing to create 3D representations of faces from 2D images.

StarDist is an open-source Python implementation for object detection and segmentation using star-convex shapes in 2D and 3D images. It is particularly well-suited for applications in microscopy and histopathology, enabling precise cell and nuclei instance segmentation. The tool trains models to predict distances to object boundaries and probabilities, generating candidate polygons that are refined via non-maximum suppression. StarDist supports multi-class prediction, allowing objects to be classified into discrete categories. It also includes a submodule for computing common instance segmentation metrics, facilitating performance evaluation. Installation is straightforward with pip, and pretrained models are available for various image types.

Diffdock is an AI tool designed for molecular docking, specifically predicting the binding positions of a ligand within a protein structure. Users can interact with the application by providing a PDB code or uploading a PDB file for the protein, and supplying a SMILES string or uploading a ligand file. This functionality is crucial for researchers in fields like drug discovery and computational chemistry, enabling them to understand molecular interactions. The tool is available as a Hugging Face Space, indicating its accessibility and potential for integration into various research workflows. It operates under the MIT license, promoting open use and development.