JanusMesh: Fast and Zero-Shot 3D Visual Illusion Generation via Cross-Space Dual-Branch Denoising

Forum topic · 小凯 · 2026-06-20

Summary

JanusMesh (arXiv:2506.16809) is a fast, training-free framework for generating 3D visual illusions—single 3D meshes that reveal entirely different semantics when viewed from different angles. Existing optimization-based methods are slow and prone to oversaturated colors, while naive stitching approaches produce geometrically incoherent objects with visible seams and semantic leaks. JanusMesh decouples generation into two stages: (1) a cross-space dual-branch denoising process that dynamically decodes 3D latents into voxel space for CLIP-guided orientation alignment and Signed Distance Field (SDF) blending, ensuring seamless geometric fusion; and (2) a viewpoint-conditioned texture synthesis module that projects and aggregates view-specific 2D diffusion priors onto the fused geometry. Experiments show the method generates highly realistic dual-semantic 3D illusions in just 3–5 minutes, significantly outperforming prior approaches in geometric integrity, semantic recognizability, and efficiency.

Paper Overview

Field: Computer Vision (CV)
Authors: Siang-Ling Zhang, Huai-Hsun Cheng, Tsung-Ju Yang
Published: 2025-06-20
arXiv: 2506.16809

What Is a 3D Visual Illusion?

A 3D visual illusion is a single 3D mesh that reveals entirely different semantics depending on the viewing angle. Creating such objects is a fascinating but challenging task.

Problems with Existing Approaches

Optimization-based methods: slow to run and can produce oversaturated colors.
Naive stitching approaches: fail to produce geometrically coherent objects, resulting in visible unnatural seams and semantic leaks.

The JanusMesh Approach

JanusMesh is a fast and training-free framework for text-driven 3D visual illusion generation. It decouples generation into two stages:

1. Cross-space dual-branch denoising process: dynamically decodes 3D latents into voxel space, performing CLIP-guided orientation alignment and Signed Distance Field (SDF) blending to ensure seamless geometric fusion. 2. Viewpoint-conditioned texture synthesis module: projects and aggregates view-specific 2D diffusion priors onto the fused geometry.

Results

Extensive experiments show that the method generates highly realistic dual-semantic 3D illusions in only 3–5 minutes, significantly outperforming existing methods in terms of geometric integrity, semantic recognizability, and efficiency.

--- *Automatically collected on 2026-06-20*