PoseCrafter: Extreme Pose Estimation with Hybrid Video Synthesis
This work addresses a critical challenge in 3D vision for applications like robotics and AR/VR, but it is incremental as it builds on existing video synthesis and pose estimation methods.
The paper tackles the problem of pairwise camera pose estimation from sparsely overlapping image pairs by proposing PoseCrafter, which uses hybrid video synthesis to generate clearer intermediate frames and a feature matching selector to improve frame selection, resulting in enhanced pose estimation performance on datasets like Cambridge Landmarks and ScanNet, especially for cases with small or no overlap.
Pairwise camera pose estimation from sparsely overlapping image pairs remains a critical and unsolved challenge in 3D vision. Most existing methods struggle with image pairs that have small or no overlap. Recent approaches attempt to address this by synthesizing intermediate frames using video interpolation and selecting key frames via a self-consistency score. However, the generated frames are often blurry due to small overlap inputs, and the selection strategies are slow and not explicitly aligned with pose estimation. To solve these cases, we propose Hybrid Video Generation (HVG) to synthesize clearer intermediate frames by coupling a video interpolation model with a pose-conditioned novel view synthesis model, where we also propose a Feature Matching Selector (FMS) based on feature correspondence to select intermediate frames appropriate for pose estimation from the synthesized results. Extensive experiments on Cambridge Landmarks, ScanNet, DL3DV-10K, and NAVI demonstrate that, compared to existing SOTA methods, PoseCrafter can obviously enhance the pose estimation performances, especially on examples with small or no overlap.