onformative trains AI to sculpt 3D models from a cube of voxels

what happens when ai becomes the creator?

In the age of technology, inspired by the rising tide of AI powered design and its controversial impact on human creators, on formative’s ‘AI Sculpting’ question: what happens when AI becomes the creator? How can human creativity and machine-made production feed each other?

As AI is increasingly adapted in design, our role as the human creator is changing rapidly. Machines, which are much faster at learning and reproducing visual results than humans ever could, are driven by efficiency, while human creators are mostly motivated by curiosity. Investigating this assemblage, the Berlin-based design studio was curious to seek an equilibrium between two co-creators – human and machine – and started an AI research project to program, observe and reinterpret the evolution of a machine that learns how to shape a 3D model. “The core of this co-creation process is that we – to a certain extent – let go of control,” notes the team.

Their experiment, presented as a curated series of digital artwork in a culminating 3D-generated exhibition environment, the journey maps to a simple cube that transforms into an increasingly recognizable shape with each iteration, eventually taking the shape of a human form.

AI learns to sculpt 3D models from a cube of voxels, using onformative
all images with permission to informative

onformative trains Ai to transform a voxel cube into a statue

Throughout history, man has had a great interest in manipulating the objects and materials of the moment for aesthetic purposes. To transfer this curiosity to the age of technology and artificial intelligence, design studio onformative developed a machine learning process and AI model as an emblematic traditional sculpting tool. ‘To best experience AI as the creator and to learn from it became our main task,’ notes the team. AI developed different strategies that seek constant improvement on the way to materialize a given form. By feeding it different tools, rules and rewards through reinforcement learning, the team was able to control the process – but not the outcome – and reveal unpredictable final shapes.

Set to achieve the given goal of sculpting a 3D model, the AI ​​was trained through reinforcement learning based on rewards and punishments. The agent, defined as a particular machine learning model, was programmed to seek the maximum reward in a trial and error process. In the voxel-based environment – ​​which started out as one big cube – infinite data was provided and a clear reward structure for the agent to move through. From this initial structure, the agent needed to remove mass to get closer to a predefined goal state.

AI learns to sculpt 3D models from a cube of voxels, using onformative

With each step, the agent is able to decide where to go, whether to remove a mass of voxels around itself, and how. To enable the learning, it was conditioned in a certain way: when extraneous mass was removed it was rewarded and when mass that should be part of the final sculpture was removed it was punished. Through trial and error, the agent created its own strategies to achieve the desired shape.

A very striking result of ‘AI Sculpting’ was the enormous parallelism of the training process, where the AI ​​was able to perform many training sets simultaneously, resulting in a large variety of sculpting results. By observing the machine’s development of the learning curve, onformative noted strategies, behavior and visual results, and continuously experimented with different parameters and predefined rules.

AI learns to sculpt 3D models from a cube of voxels, using onformative

through the eyes of the machine: Visualizing AI

By exploring the concept of co-creation, onformative decided to take on the view of the agent to see the process from the perspective of the creator. In an attempt to interpret the agent’s decision-making, the design team experimented with visualizing AI data as confidence, or punishment and reward for individual steps in the 3D environment. By highlighting the agent’s path through the block, for example, they gained a different perspective on the creation process itself.

Finally, the team experimented with their creative expression by implementing different tools for the agent to choose from. Through the reinforcement learning setup, this process allowed for more complexity and resulted in a variety and depth of strategies as well as visual outcomes. By looking at the traces of different tools on the surface of the new forms, the sculpting process becomes visible. Different tools have their own unique fingerprint – from coarse to fine and fast to slow – each decision the agent made by choosing a different tool or mode of orientation had a different visual result and impact. AI learns to sculpt 3D models from a cube of voxels, using onformative AI learns to sculpt 3D models from a cube of voxels, using onformative

artificial intelligence learns to sculpt 3D models from a cube of voxels, using onformative

AI learns to sculpt 3D models from a cube of voxels, using onformative

Leave a Reply

Your email address will not be published. Required fields are marked *