1 paper
Yiqing Liang, Eliot Laidlaw, Alexander Meyerowitz +2
From video, we reconstruct a neural volume that captures time-varying color, density, scene flow, semantics, and attention information. The semantics and attention let us identify…