FurE: Efficient Instance-Specific 3D Fur Reconstruction without Animal-Fur Datasets
FurE reconstructs editable 3D animal fur from multi-view images without training on animal-fur datasets.
The method builds a strand-based groom by optimizing a root-conditioned latent field and decoding strands with a PCA model learned from human-hair data. It also reconstructs a defurred animal body using local fur-thickness cues and part-based priors. The authors report a 10x speedup over dense per-strand optimization while retaining strand fidelity across synthetic and real-world sequences. ArXiv · AI/CL/LG's note
The method builds a strand-based groom by optimizing a root-conditioned latent field and decoding strands with a PCA model learned from human-hair data. It also reconstructs a defurred animal body using local fur-thickness cues and part-based priors. The authors report a 10x speedup over dense per-strand optimization while retaining strand fidelity across synthetic and real-world sequences. ArXiv · AI/CL/LG's note
score 4