VidMap: Exploiting temporal structure for video-based Structure-from-Motion

Loading video
Loading videoVidMap brings the temporal structure of SLAM into global Structure-from-Motion for accurate 3D reconstruction of long, unconstrained videos (ECCV 2026).
Category: perception
Author: @zador_pataki
Date: 2026-08-07T00:00:00
Duration: 52.93s
Reference: https://arxiv.org/abs/2607.27194





