Blank {Compose}™ is a custom design and strategy solution that uses Deep Learning to make sense of 3D data by building a solution space.
Generate new designs using semantic controls. Explore the infinite solution space yourself or get inspired by random walks.
Analyze a shape in relation to all others, be it pre-existing or newly generated. Get qualitative and quantitative measures.
Take a random walk through the solution space and watch Blank {Compose} dream up infinite new forms.
Freeze any moment to set it as a starting point for the ensuing exploration.
We have developed a data pipeline that automatically infers a solution space from thousands of 3D models and qualified metadata. A set of comprehensive tools enables the harvesting of this space.
Use your custom data labels to create infinite, targeted variants of your solution.
Explore how high-level properties translate to concrete shapes.
Blend seamlessly between any given solution. Start from your own design or retrieve existing market products.
This offers “semantic” shape blending for the first time, where every intermediate step represents a valid solution.
Understand how your design performs based on its location in the solution space.
Instantly evaluate how your solution relates to existing ones, and how it measures against your defined labels.
Objectify the design discussion by controlling implicit shape properties.
Blank {Compose} uncovers emergent directions in the solution space that translate to subjective impressions.
Subscribe if you'd like to be the first to know about new Blank {Compose} developments.