Find open source projects actively accepting contributors. Search repositories, filter by program milestones, difficulty tags, or tech stack.
Use our Orbit AI Matcher to find out! Get instant matching scores based on your developer skills, preferred frameworks, and contribution experience.
Convert your selected open-source project into a winning GSoC, LFX, or Outreachy application using Proposal Studio.
<p>Extracting metadata from AV1 videos is very useful for codec research. This metadata can be used to create a video analyzer, analyze coding performance of a video, encoder development and many other purposes. While there is a way to extract metadata by using the reference decoder, libaom, the extraction of sophisticated metadata is lacking.</p> <p>This project succeeded in adding additional metadata extraction APIs to libaom and visualizations of AV1 bitstream features in the AOMAnalyzer. The features added involved analyzing compound prediction types, motion modes and a full-featured debugging tool for analyzing and visualizing film grain, a coding tool unique to AV1.</p> <p>This project culminated with a presentation to the AOMedia working group. These industry professionals are the core developers who were responsible for the development, testing and standardization of the new AV1 video format. A highlight of my GSoC experience was getting feedback directly from the authors of the coding tools I worked to visualize.</p>
<p>digiKam 7.2 as of now, is fully compatible with Qt 5.15.2 on all major platforms. Qt6 has been released recently. Given that Qt5 and Qt6 are 8 years apart, the way of doing things has changed and complies with the C++17 standard. This project is aimed at preparing the digiKam project to be ported and released with Qt6 on Linux. Implementations that rely on deprecated, removed, improved APIs will be ported with a lot of regression testing.</p> <p>The AppImage is known to be broken as it doesn’t support internationalization, or proper ICU support. ICU works fine on other sources for digiKam like flatpak and native binaries. This issue will be fixed by porting the AppImage builder scripts to use Qt 5.15.2 with KDE Patches collection (Not for the moment) which was recently announced (<a href="https://dot.kde.org/2021/04/06/announcing-kdes-qt-5-patch-collection" target="_blank">https://dot.kde.org/2021/04/06/announcing-kdes-qt-5-patch-collection</a>). New scripts will be implemented to build the AppImage with Qt6 in future.</p>
<p>DFFML is a machine learning based project which provides APIs for training and testing datasets using various machine learning frameworks such as scikit-learn, tensorflow and more, making it easy to generate datasets, train and use machine learning models, and integrate machine learning into new or existing applications.</p> <p>At present, there is no way of training and testing the existing machine learning models in DFFML on image datasets, so I have selected 2 image processing python libraries OpenCV and Scikit-Image to wrap in DFFML.</p> <p>The project is divided into 2 parts:</p> <ol> <li>Wrapping the Image Processing Libraries namely OpenCV and Scikit-Image.</li> <li>High Level Operation Workflow, i.e., Custom Operations which will act as high level operations implementing a predefined flow of OpenCV and Scikit-Image functions.</li> </ol>
<p>Gravitational lensing has been a cornerstone in many cosmology experiments, and studies since it was discussed in Einstein’s calculations back in 1936 and discovered in 1979, and one area of particular interest is the study of dark matter via substructure in strong lensing images. While statistical and supervised machine learning algorithms have been implemented for this task, the potential of unsupervised deep learning algorithms is yet to be explored and could prove to be crucial in the analysis of LSST data. The primary aim of this GSoC 2020 project is to design a python-based framework for implementing unsupervised deep learning architectures to study strong lensing images.</p> <p>Refer to the <a href="https://arxiv.org/abs/2008.12731" target="_blank">paper</a> for more details.</p>
<p>One of the main ingredients in DOLFIN’s native support of parallel computations is the mesh partitioner. The mesh partitioner seeks to ensure a load balance among processing elements and reduce the number of shared elements on partition boundaries, in order to minimize the communication overhead. Currently, the users can pick from SCOTCH or ParMETIS partitioners, neither of which are actively maintained. The KaHIP partitioner is an alternative that is being actively maintained and promises more scalable and higher quality results than the state-of-the-art partitioners such as ParMetis or PT-Scotch. The purpose of this project is to add the KaHIP partitioner to DOLFIN's graph wrappers and mesh partitioning and investigate whether the promised improvements will reflect on the dolfin's parallel toolchain.</p>
<h3>Implementation of Newsfeed (a social feature) in Rocket.Chat Server</h3> <p>This project adds a newsfeed for every user. In it, the users can follow other users. Whenever a user posts something in a public channel or does any public activity, it would appear in the newsfeed of all his followers with a link to the original post in the original channel. This would greatly help in discovering new conversations, increasing accountability of people in the organisation & increasing many folds the time a user spends on RC</p> <p><strong>The most challenging part of creating the newsfeed is to make it scalable to support a large number of concurrent users (by using an adaptive follower/following relation model) and to benchmark the server performance in relaying the correct messages with filters. This is being done by using proper schema and proper database queries. Messages in newsfeed are fetched in real-time from messages stored for other channels.</strong></p> <p><em>Key Features:</em></p> <ul> <li>Support for Rich Message blocks</li> <li>Auto-summarization of messages to prevent spam in newsfeed</li> <li>Expandable embedded conversations</li> <li>Upvoting System & Trending messages</li> <li>Group messages by origin user, date, channel etc.</li> <li>Integration with other projects</li> </ul>
<p>LabelLab is an image analyzing and classification platform. The web application should allow users to upload batches of images and classify them with labels. It will also have the features to run classifications against a trained model. LabelLab also has a user project management component as well as an image analyzing component.</p>
<p>The aim of this project is to finish incomplete implementations of the algorithms in aima-python, and refactor some of the existing ones that do not work as intended. It also aims to add tests for these algorithms and increase the coverage of the existing ones. Detailed jupyter notebooks and visualizations will be written explaining the intuition behind the concepts, the functioning and the usage of the algorithms with real-world examples.</p>
<p>The general goal of this project is to provide perfect user experience for users of Performance Co-Pilot on Windows. Now Windows port of Performance Co-Pilot can be built on Fedora Linux. That requires configurated Fedora operating system with all necessary packages. Additionally, users must be experienced in building from source and packaging. All mentioned reduces scalability, integrity and use cases of PCP framework. Users may try to find relatively easier ways to gain their demands. Additionally, there are problems with sending patches and updates to users which can potentially cause even security risks. So, there is no need in further explanation of the utility of this project.</p>
<p>Allow any user to keep the changes he does when he's booted into the live system across reboots.</p>
<p>Virtual reality has become a rage with Oculus Rift VR headsets and Google Daydream View headsets hitting the markets. More importantly, with the introduction of google cardboard, even a smartphone can now be used to visualize stereoscopic images and videos for literally no extra cost. This has potential applications in the field of education as well. The idea is to extend 3dmol.js to be able to project stereoscopic images so that it can be viewed using a phone browser through any low cost cardboard headset. Stereoscopic rendering will make the visualization of simulation results much clearer, especially while explaining interesting atomic scale phenomena such as bonds breaking or denaturation of proteins etc. to students in class. The users can download any of their simulation data using 3dmol.js, which will convert their files to a pair of side-by-side images providing a stereoscopic effect.</p>
<p>The OpenWorm project is building a simulation of the C. Elegans in an open science fashion. The model completion dashboard is a web-based visualization of the digital versions of biological entities that are currently captured within OpenWorm’s database API, PyOpenWorm. This interface is designed to display the results of the unifying modeling activity, and should be coordinated with the crowdsourcing platform for C. Elegans ion channels known as ChannelWorm. This interface allows a user to drill down into OpenWorms' model, and view the states of completion of modeled components at each level.</p>
<p>My perception for the next evolution of Cividroid revolves around the concept of ‘Dashlets’. During my frequent visit to the CiviCRM Dashboard on the demo set-up, I find that, it gives a holistic view of many of the crucial data one would want to see from their CiviCRM installation. Since they’re mostly powered by CiviReports, there is an opportunity to derive a lot of new Dashlets. Introducing Dashlets into Cividroid sounds very natural and they will feel just like home on Android which itself has a concept of ‘widgets’.</p>
<p>GNOME Calendar is the calendar application for GNOME. Although it is a usable product, it still misses the important Week and Day view required by users for organizing the week, and planning a given day. The project aims at adding the feature of week-view in the calendar for better user experience for an end user.</p>
<p>Fingerprinting for Programs is aimed at analyzing code blocks on a semantic level. This is done by symbolically executing the code block via SPF over JPF in order to generate path constraints for each possible path that a program may take. The path constraints are then put into canonical form and analyzed. Level of refinement of the path constraints dictate the level at which we model programs i.e. If we do not refine the path constraints at all, the program is modeled on a syntactical level. If we stop refinement early, the program is modeled on a semantic level but syntactical structure of the program is still considered. If we completely refine the path constraints and put them into canonical form, the program is modeled on a semantic level with no consideration towards the syntactical structure of the program. This gives us options on applications of fingerprinting for programs. Further research in this field could result in semantic patterns not just being observed but rather analyzed. Applications of this analysis include the possible creation of a semantic auto-complete for programmers.</p>
<p><strong>Thread Scheduling in HPX</strong><br> HPX currently supports seven types of schedulers:</p> <ul> <li>Priority Local Scheduling </li> <li>Static Priority Scheduling </li> <li>Local Scheduling </li> <li>Static Scheduling </li> <li>Priority ABP Scheduling </li> <li>Hierarchy Scheduling </li> <li>Periodic Priority Scheduling </li> </ul> <p>All of these schedulers are currently statically linked and must be specified at compile-time. The aim of the project is to make these dynamically loaded instead. These will be converted to the plugin system already present in HPX, which is used by message-handlers and parcelports. Factory design pattern will be followed. A scheduler-factory-base class will provide the template from which the factories of the individual schedulers are derived. These will then return scheduler-objects which can be used by executors.</p>
Problem: Imbalanced datasets pose a significant challenge in machine learning, where standard metrics like accuracy can falsely evaluate a model that simply predicts the dominant class. Currently, the R community lacks a comprehensive and unified framework for handling imbalanced data and X-Risks problems natively integrated into the mlr3/mlr3torch ecosystem. Solution: This project, mlr3torchAUM, provides a comprehensive solution for imbalanced classification in R. It will bridge the current gap by integrating state-of-the-art loss functions, advanced evaluation metrics, and various data sampling techniques directly into the mlr3/mlr3torch pipeline. Deliverables: 1. State-of-the-Art Loss Functions: Implementation of algorithms including log-linear all-pairs squared hinge loss, Area Under Min (AUM) loss, and cost-sensitive logistic loss. 2. Sampling Operators: A robust suite of over-sampling techniques including SMOTE and ADASYN, alongside under-sampling techniques, all implemented safely under a `lazy_tensor` mechanism to avoid memory issues. 3. Advanced Evaluation Metrics: Integration of Precision-Recall curves, Macro/Micro AUM/AUC, and MCP/IMCP to fully support multiclass imbalanced scenarios. 4. LibAUC Integration: Porting cutting-edge losses and optimizers from the Python LibAUC library into R. 5. Engineering & Release: Comprehensive Test Driven Development, GitHub Actions for CI/CD, detailed vignettes, and a final submission to CRAN.
This project will build a native printer setup tool for the COSMIC desktop. It will add a Printers page in Settings and an Add Printer flow for legacy or specialty devices. The goal is to give COSMIC a modern printer management experience that matches the move toward IPP and Printer Applications in the Linux printing stack. I plan to implement it in Rust inside the cosmic-settings workspace, with a backend layer that keeps the UI stable as CUPS and Rust bindings evolve. The main view will show and group printer destinations, display status and default-printer information, and provide access to device web interfaces. The Add Printer flow will focus on finding legacy devices, suggesting suitable Printer Applications, and refreshing the main view after setup. Deliverables include a working Printers page, grouping and de-duplication logic, web-interface actions, an Add Printer workflow, tests, and documentation.
MLIR already makes allocation and deallocation structure explicit after bufferization and the ownership-based deallocation pipeline, but it still lacks an upstream static memory planner that can analyze buffer lifetimes and safely reuse memory for well-structured workloads. This is especially valuable for structured ML, HPC, and accelerator-oriented compilation flows where dynamic allocation is expensive or undesirable. I propose to build this in two incremental steps. First, I will implement an analysis pass that identifies eligible memref.alloc / memref.dealloc patterns, performs alias-aware lifetime analysis, computes planning metadata such as size and alignment, and reports reusable candidates and skip reasons. Second, I will build a conservative rewrite pass that rewrites eligible dynamic allocations into memref.subview-based slices of one or more statically allocated arena memrefs for legality-gated v1 cases. The main deliverables are: an upstream analysis pass for static memory planning, a conservative v1 rewrite pass for same-block structured cases, memory-space-aware planning support, strong lit/FileCheck coverage, documentation of supported and unsupported cases, and an incremental PR series that is realistic to review and land upstream.
Strong gravitational lensing is a powerful probe of dark matter structure and cosmology, yet no foundation model exists that is purpose-built for lensing data with theoretical training guarantees. This proposal introduces WaveLens-JEPA, a self-supervised vision foundation model for gravitational lensing that extends LeJEPA (Balestriero & LeCun, 2025) to the astrophysical domain. The core idea is to decompose lensing images using a 2D Discrete Wavelet Transform before encoding, separating the physically meaningful frequency scales — Einstein ring morphology, subhalo perturbations, and ALP vortex structure — into orthogonal channels. A physics cross-attention module injects convergence-map priors from the lensing equation at every ViT block, while Einstein-Ring-Aware Masking (ERAM) targets the physically informative annular region during pre-training. Fourier phase regularization preserves substructure-discriminative spatial information that standard encoders discard. The training objective inherits provable collapse-free guarantees via Sketched Isotropic Gaussian Regularization (SIGReg), unlike all prior lensing SSL methods. The pre-trained backbone is evaluated on six downstream tasks: two substructure classification setups, lens finding, mass density regression, super-resolution, and a novel interpretability task via Wavelet Attribution Maps (WAM). Deliverables include: (1) open-source modular codebase, (2) pre-trained backbone weights on Hugging Face Hub, (3) a six-task evaluation benchmark, (4) a sim-to-real domain adaptation protocol targeting COSMOS HST data, and (5) a full research paper.
This project aims to integrate bytetrack algorithm into a Mediapipe graph using OpenVINO inference calculators. The bytetrack graph will be deployed on OpenVINO model server to perform efficient object tracking on a video input. The deliverables of this project include a bytetrack graph in OpenVINO mediapipe fork with custom calculators, YOLOv10 model and bytetrack graph packaged in OpenVINO model server, and a Python client for the bytetrack graph to perform bytetrack algorithm on video stream. The final phase of this project includes evaluation of bytetrack on metrics MOTA, IDF1, and HOTA. The goal of this project is to bring the bytetrack algorithm to OpenVINO model server to leverage its efficiency, scalability, and robustness for production-grade inference.
This project aims to improve the Debian packaging process for FOSSology, an open-source license compliance tool. Currently, FOSSology relies on CPack for packaging, which has limitations in flexibility and Debian policy compliance. This proposal suggests transitioning to Debian's native packaging tools (dpkg, debhelper, and lintian), automating the build process, and integrating continuous integration (CI) for better package management. Deliverables include structured Debian packages, automated build scripts, CI pipelines, and detailed documentation to facilitate future maintenance. This work will make FOSSology more accessible, easier to install, and compliant with Debian standards, benefiting system administrators and developers.
The current End-to-End (E2E) test coverage for the new React frontend stands at approximately 30%, leaving critical functionalities untested. This project aims to enhance E2E test coverage to at least 80%, ensuring comprehensive validation of the entire application workflow. By identifying gaps, developing additional test cases, optimizing execution, and integrating tests into the CI/CD pipeline, this initiative will significantly improve system reliability, user experience, and overall software stability.
This project aims to overcome the limitations of the current Build Script system by allowing multiple, well-organised build scripts, that are optional and run only when required. This approach will require less auditing, will enhance the build transparency and reduce the compile-time overhead. Build scripts in Rust facilitate compiling and linking 3rd party code, such as existing C/C++ libraries, before building a package. They are immensely important for interoperability since it prevents replacing/rewriting the already existing tools that are well-optimized for their tasks. However, the current build system has certain limitations. Only one build script is allowed per package, and it must be run before building the package, often causing unnecessary computation (e.g., during unit tests). Single Large script needs extra auditing, increase compile-time, and reduce transparency for third-party build tools, especially while passing parameters. They also cannot run unit tests within themselves. This project aims to overcome the above limitations by allowing multiple, well-organized build scripts that can be run directly from the Cargo.toml. These scripts are optional and run only when required, avoiding unnecessary expensive logic. With this, only the shared build scripts will need auditing. This approach enhances build transparency, reduces compile-time overhead, and even enables unit testing of build scripts if they are treated as a dedicated package.