MIT's Meschers Tool Visualizes & Edits Impossible Objects
Researchers at MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) have developed a novel tool called “Meschers” that enables the visualization and manipulation of “impossible objects” — geometries that defy the laws of physics, reminiscent of the mind-bending artwork by M.C. Escher.
Traditionally, computer graphics scientists and designers can only recreate these optical illusions in 3D through physical workarounds, such as bending or cutting a real shape and positioning it at a precise angle. This approach has significant limitations: altering the smoothness or lighting of the structure reveals its true, physically possible nature, rendering it unsuitable for accurate geometric analysis.
Meschers offers a unique solution by converting images and 3D models into “2.5-dimensional” structures. This innovative representation models how humans perceive these impossible shapes. Instead of forcing global consistency, the tool focuses on the “local consistency” of an object’s parts – meaning individual sections appear plausible, even if the whole cannot exist in reality. Behind the scenes, Meschers works by knowing the image’s x and y coordinates, along with the differences in depth (z coordinates) between neighboring pixels. This allows the tool to indirectly reason about and manipulate impossible objects. Crucially, it enables users to relight, smooth, and study these unique geometries while preserving their optical illusion.
The versatility of Meschers opens new avenues across several fields. For geometry researchers, it could facilitate complex calculations like “geodesics” (the shortest distance between two points on a curved impossible surface) and simulate “heat diffusion” (how heat dissipates across such an object). Artists and computer graphics scientists can now explore an entirely new class of shapes, creating physics-defying designs in multiple dimensions. Additionally, perception scientists could use the tool to better understand the precise point at which an object transitions from plausible to truly impossible in human perception.
Ana Dodik, lead author and an MIT PhD student, expressed her aim to design computer graphics tools that transcend the replication of reality. “Using Meschers, we’ve unlocked a new class of shapes for artists to work with on the computer,” she stated. Senior author Justin Solomon, an associate professor at CSAIL, added, “Meschers demonstrates how computer graphics tools don’t have to be constrained by the rules of physical reality. Incredibly, artists using Meschers can reason about shapes that we will never find in the real world.”
The team has demonstrated the tool’s capabilities on various impossible objects, including an “impossibagel” (a bagel shaded in a physically impossible way) to simulate heat diffusion, and a model of a dog on a skateboard to showcase its ability to tweak shading while maintaining the optical illusion. Meschers can also subdivide these structures for more precise calculations, reduce visual imperfections, and even convert drawings and images of impossible objects into high-dimensional designs through “inverse rendering.”
The research will be presented at the upcoming SIGGRAPH conference in August. Looking ahead, Dodik and her colleagues plan to develop a more user-friendly interface, build more elaborate scenes, and explore broader applications in collaboration with perception scientists.