1 paper · 1 filter
Yufei Cai, Xuesong Niu, Hao Lu +3
Current visual generation models are capable of producing high-quality content, yet they lack a coherent perception of the spatial structure. Existing generative novel view synthes…