Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

SF3D: Stable Fast 3D Mesh Reconstruction with UV-unwrapping and Illumination Disentanglement

About

We present SF3D, a novel method for rapid and high-quality textured object mesh reconstruction from a single image in just 0.5 seconds. Unlike most existing approaches, SF3D is explicitly trained for mesh generation, incorporating a fast UV unwrapping technique that enables swift texture generation rather than relying on vertex colors. The method also learns to predict material parameters and normal maps to enhance the visual quality of the reconstructed 3D meshes. Furthermore, SF3D integrates a delighting step to effectively remove low-frequency illumination effects, ensuring that the reconstructed meshes can be easily used in novel illumination conditions. Experiments demonstrate the superior performance of SF3D over the existing techniques. Project page: https://stable-fast-3d.github.io

Mark Boss, Zixuan Huang, Aaryaman Vasishta, Varun Jampani• 2024

Related benchmarks

TaskDatasetResultRank
Single-view 3D ReconstructionGSO (test)
CD0.137
18
Reflective Object ReconstructionShiny Blender
PSNR18.03
18
3D human reconstructionCustomHuman (test)
Chamfer Distance (P2G)1.657
16
3D human reconstructionTHuman2 (test)
CD (Pred to GT)1.441
16
3D human reconstruction2K2K (test)
Chamfer Distance (Pred to GT)1.204
16
Image-to-3D GenerationGoogle Scanned Objects (GSO)
CLIP Similarity74.63
14
Novel View SynthesisGSO + Stanford ORB
PSNR17.64
13
3D ReconstructionGSO + Stanford ORB
Chamfer Distance (CD)0.132
13
Novel View SynthesisuCo3D
PSNR12.79
13
3D ReconstructionuCo3D
Chamfer Distance (CD)0.248
13
Showing 10 of 35 rows

Other info

Code

Follow for update