Dai et al.
BEV maps enable human-friendly control for embodied navigation video generation
Map2Vid reformulates navigation video synthesis as a BEV-map-to-video problem, providing intuitive spatial grounding and a large-scale benchmark.
Chinese Tech
Yulin Dai · Yuqian Fu · Tianwen Qian · Zhen Xing · Bojia Zi · Xiaoling Wang
East China Normal University · King Abdullah University of Science and Technology (KAUST) · Alibaba Group · The Chinese University of Hong Kong
Research Digest··3 min read
The authors introduce Map2Vid, a task that generates egocentric navigation videos conditioned on bird's-eye-view (BEV) maps, offering a more intuitive alternative to text prompts or camera trajectories.
§