AR Overlay: Training Image Pose Estimation on Curved Surface in a Synthetic Way
This work addresses a domain-specific challenge in AR by improving multi-image detection for curved surfaces, though it appears incremental as it builds on existing methods for known bottlenecks.
The paper tackles the problem of pose estimation for curved images in spatial computing by proposing a pipeline that detects multiple logo images simultaneously using only original images as input, enabling more effects in AR applications.
In the field of spatial computing, one of the most essential tasks is the pose estimation of 3D objects. While rigid transformations of arbitrary 3D objects are relatively hard to detect due to varying environment introducing factors like insufficient lighting or even occlusion, objects with pre-defined shapes are often easy to track, leveraging geometric constraints. Curved images, with flexible dimensions but a confined shape, are essential shapes often targeted in 3D tracking. Traditionally, proprietary algorithms often require specific curvature measures as the input along with the original flattened images to enable pose estimation for a single image target. In this paper, we propose a pipeline that can detect several logo images simultaneously and only requires the original images as the input, unlocking more effects in downstream fields such as Augmented Reality (AR).