Cascadeur, the AI-assisted character animation software, has rolled out a new update that further refines the animation workflow. Building on the Inbetweening functionality introduced in the previous release, this update improves its capabilities while adding a diverse range of other enhancements. 

The upgraded Inbetweening feature now uses a faster, more advanced algorithm capable of generating motion between poses for up to 120 frames. It's also more context-aware, enabling seamless transitions between different animations by recognizing velocities and blending them naturally. In addition, a new dedicated tab lets users select the animation “style” they want to generate, such as walking or running, while benefiting from improved handling of tangents and transitions for smoother, more polished results.

Improved Inbetweening | Cascadeur 2025.2 Feature Highlights

Speaking of tangents, the update introduces a new tool that lets users edit animation trajectories directly in the viewport using tangents. Nekki, the company behind Cascadeur, describes this as an early-stage feature that will continue to evolve. For now, it's available exclusively in Point Controller mode.

This release brings several upgrades to the Fulcrum tools. Fulcrum points are now calculated with greater precision, allowing users to apply Cascadeur's AutoPhysics, a feature designed to analyze and adjust animations for enhanced physical accuracy, immediately without first needing to unbake the animation. 

The Fulcrum motion cleaning feature can now be applied automatically, helping to eliminate common issues such as foot sliding and knee popping, and resulting in more stable, natural movement. Additionally, refinements to automatic fulcrum detection make it especially valuable for motion capture cleanup and AutoPhysics workflows.

A new AI system can now detect fulcrum points even when a character's feet are at varying distances from the ground or sliding significantly. This eliminates the need for manual fulcrum adjustments, enabling foot cleanup in just a few clicks.

The Visual Mocap feature has been enhanced with improved recognition of both depth and the ground plane in the capture environment. The tool now automatically improves the global positioning of the character and removes foot sliding. Users will also notice several refinements to AutoPhysics, particularly in scenarios where characters interact with each other or with objects in the scene.

Other additions include a new trajectory type for visualizing orientation over time, support for the 3Dconnexion Mouse on Windows, and a range of general improvements, including UI and UX enhancements, upgraded physics, better rigging support, and cleaner mocap imports.

For more information, be sure to check the official release notes.

Blips Game Music Packs

Note: You're leaving the Blips Blog. The visited links may be subject to different privacy and cookie policies.