Development update
Where we go from here.
AImation Studio is at a testing stage. Our priority now is to learn how it performs on more computers and in real creative projects before committing to the next stage of development.
More people. More useful testing.
We need more than 10,000 additional downloads to broaden our testing and build a more reliable picture of how AImation Studio works in practice.
Downloads are only the starting point. What matters is people actually using the application: completing captures, trying different setups, and telling us where things go wrong.
Download and take part ↗Current priority · Public testing
Phase 10: Public testing, feedback and the next development stage
Our immediate priority is to test AImation Studio with more people, more hardware configurations and real creative projects. We are aiming for 10,000+ additional downloads to reach a broader group of testers. This is a recruitment goal, not a guarantee of reliable results: completed captures, repeat use and reproducible feedback matter more than downloads alone.
What we want to test
- Getting started: installation, account creation, email confirmation, trial activation and the first successful capture. Identify unclear instructions and places where new users get stuck.
- Hardware and recording conditions: different Windows PCs, GPUs, cameras, lighting, backgrounds, camera distances and supported single- or multi-camera setups. Record the application version and configuration alongside each result.
- Motion quality: walking, turning, faster movements, partial occlusion and hand motion. Look for jitter, lost tracking, foot sliding and recovery after tracking is interrupted.
- Complete workflows: capture, cleanup in the editor, saving and reopening a project, and exporting into the tools testers actually use. Check whether the result is useful in a real animation or game project.
- Stability and performance: startup problems, crashes, longer sessions, processing time and memory use. Distinguish repeatable issues from isolated failures.
How to send useful feedback
Please include your application version, Windows version, CPU/GPU, camera setup, steps to reproduce, and the expected and actual result. Screenshots or a short sample can help when you choose to share them. Do not include passwords, license keys or private footage. Use the contact page to send a report; we are not claiming that automatic reporting or telemetry is already available.
How we will judge progress
We want to understand how many testers reach their first capture and successful export, which configurations fail, which issues recur, and whether people return to finish a project. We will use these findings to improve onboarding and tutorials and to prioritize fixes, then ask testers to check the affected workflows again. Download count alone will not mark this stage as complete.
What happens after testing
Further development depends on securing funding. The next stage will be defined around the evidence from testing and the resources available. Existing feature milestones describe development directions, not guaranteed delivery dates. We want to move forward once funding and a realistic scope are in place.
Development milestones
Our work so far and the directions we want to explore. Further development depends on testing outcomes and funding; these entries are not a delivery schedule.
Phase 1: Foundation and Camera Feed IntegrationCompleted
Established a modern and versatile application framework. Invested considerable time in optimizing the project's development process, prioritizing ease of use and developer-friendliness.
Implemented a highly efficient camera feed pipeline to seamlessly integrate image inputs from diverse sources.
Phase 2: Precision and Hardware-Independent RecordingCompleted
Establish infrastructure for the storage, processing, and retrieval of images from diverse sources.
Develop a system to estimate intrinsic camera parameters for various hardware devices.
Phase 3: Pose Estimation, Editor, and Advanced Pose ToolsCompleted
Developed infrastructure for running and reading from AI models.
Implemented base pose estimation with 33 keypoints for accurate human movement tracking.
Developed accompanying tools for efficient storage, retrieval, and manipulation of motion data.
Phase 4: Depth and User-Friendly GUICompleted
Implemented robust stereo calibration for accurate extrinsic parameters estimation and 3D reconstruction.
Create a basic GUI for system calibration and motion capture recording.
Enable multi-device recording.
Phase 5: GPU Computing and Enhanced PerformanceCompleted
Implemented the required infrastructure and code to efficiently run the AI model on Windows GPUs. It now achieves an impressive 23 frames per second, providing nearly real-time output on a standard PC ( NVIDIA GTX 1060 )
Substantial progress has been made in writing the essential code for the 3D editor.
Phase 6: Hands Estimation, Face Estimation, Networking Backend, and Multi-View Multi-Screen 3D ViewportCompleted
Expanded the pose model to include an additional 42 keypoints, with 21 keypoints dedicated to each hand.
Established the foundation for a gesture recognition system.
Incorporated the face mesh AI model into the system, although it is currently inactive due to the additional processing requirements for face mesh data.
Created a networking backend to enable smooth connections between network devices and AIMation.
Developed a fully operational multi-view, multi-screen 3D viewport that enables users to observe the animation from different angles.
Phase 7: Mobile Application DevelopmentCompleted
The required protocol for remote AIMation-studio control has been successfully implemented.
Begin developing a mobile application for iOS and Android platforms, focusing on AImation control and image streaming.
Phase 8: Refining Animation Output & Editor ExperienceCompleted
Implemented advanced post-processing tools to achieve smooth and precise animations.
Developed an FBX export tool for easy transfer of animations to 3D software platforms.
Enhanced the 3D editor and user interface for improved usability.
Phase 9: Advanced Hand Detection and Motion Capture FiltersIn progress
Enhanced hand detection capabilities.
Initiate the development of additional calibration methods for mobile phones to estimate depth using a single sensor, enhancing the depth perception capabilities of the system.
Begin the implementation of smoothing splines and butterworth filters to smooth and polish motion data
Phase 11: Integration with Game Engines, Custom Streaming and more export formatsIn progress
Integrate Animation Studio with popular animation tools such as Blender and Maya, as well as game engines like Unity and Unreal, allow users to stream their animations to custom clients.
Implement industry-standard exporters for motion formats.
Phase 12: Additional animation tooling & face captureIn progress
Implement graph-based animation retargeting system.
Implement advanced inverse kinematics solver with constraints.
Integrate face mesh functionality into estimation pipeline.
Implement dedicated face recording window with additional tooling for face recording sessions.
Phase 13: Additional device supportIn progress
Expand our device support by incorporating action cameras, other high-quality cameras as well as AI cameras.
At present, we already offer support for OpenCV depth cameras and their corresponding API.
Phase 14: Gesture Recognition SystemPlanned
Integrate modular gesture recognition AI model.
Enhance animation output with recognized gestures.
Allow user customization of gesture sets.
Phase 15: Programmable ActionsPlanned
Enable programmable actions based on recognized gestures.
Implement user-friendly interface for action creation.
Phase 16: API for Game IntegrationPlanned
Develop an efficient and robust API for VR/AR games to enable seamless real-time avatar control.
Release AImation as a high-performance API, empowering game developers to leverage its capabilities.
Moreover, integrate AImation seamlessly with AR goggles to enhance the experience of real-time business meetings conducted over the web
Phase 17: Enable Mobile GPU ComputationPlanned
Allow our models to run on mobile devices since they are compact enough.
Implement the extraction of depth data from depth sensors on iOS.
Implement depth using the multi-camera API on Android, considering that most new phones have a multi-camera setup. Additionally, research additional techniques for depth extraction on smartphones.
The mobile team is responsible for handling these tasks, and they are conducted in parallel with AImation-studio development.
FuturePlanned
Enhance the system by introducing the capability to substitute pose estimation AI models with object feature extractor AI models. This new functionality will enable the recording and reconstruction of objects within a 3D space.
Research and develop additional tooling and AI models with a strong focus on single device depth perception and multi-device depth perception.
Explore potential integrations into fitness-related markets, with the goal of leveraging the technology for applications such as personalized workout tracking, posture correction, and gesture-based exercise guidance.
Investigate potential integrations into DIY security systems, aiming to enhance surveillance capabilities, object recognition, and anomaly detection to provide users with a more comprehensive and intelligent home security solution.
Explore opportunities for integration into the healthcare industry, including remote patient monitoring, fall detection, and gesture-based control for medical devices.
Explore possibilities for integrating the technology into augmented reality (AR) and virtual reality (VR) applications, with the aim of enhancing immersive experiences and interactions within these environments.
Investigate opportunities for integration into the healthcare industry, exploring applications such as remote patient monitoring, fall detection, and gesture-based control for medical devices.
Continuously conduct research and development efforts to expand the capabilities and explore new potential applications of the technology. Collaborate with industry partners and experts to identify market opportunities and optimize integrations
Help us understand what works.
If you try AImation Studio, tell us about your hardware, what you were trying to capture, and what happened. If you’d like to discuss funding or a partnership, we’d like to hear from you too.
Try Aimation Studio with your own recording.
Try the workflow with your own footage. Then choose the license that fits your project.