At its core, Vision utilizes state-of-the-art visual models to analyze footage. It can detect a wide range of common objects within a scene—from people and vehicles to animals and everyday items. The plugin offers multiple detection modes, including specialized models for faces, people, and vehicles, allowing users to fine-tune what the plugin focuses on.
The v1.0.1 update brings essential stability improvements and optimizations to the core Vision technology 1.2.1.
: Users can customize fonts, rename object labels, and choose which specific objects are highlighted. Aescripts Vision v1.0.1.zip
Generates sci-fi-inspired UI elements, such as bounding boxes and network lines, that track with identified objects.
Added support for RTX 50xx GPUs and shifted requirements to After Effects 2024 or newer. At its core, Vision utilizes state-of-the-art visual models
While the initial launch brought the "wow" factor, focuses on stability and workflow refinement. The main highlight of this update is a critical bug fix that improves reliability during the detection process, ensuring your creative flow isn't interrupted by unexpected errors. Key Features to Explore
What (e.g., rotocloning, depth effects, or graphics tracking) are you trying to accomplish first? Share public link The v1
Aescripts Vision v1.0.1 has a wide range of potential applications across various industries, including:
★★★★☆ (4/5) Recommended for: Character animators, riggers, and anyone working with heavy, multi-layered compositions.
The actual compiled code that lives in the host application's Plug-ins folder.
It can instantly identify and isolate over 80 common object categories (people, cars, electronics, etc.) using machine learning.