Streamlining the Image Stitching Pipeline: Integrating Fusion and Rectangling into a Unified Model

arXiv:2404.14951v1 Announce Type: new
Abstract: Learning-based image stitching techniques typically involve three distinct stages: registration, fusion, and rectangling. These stages are often performed sequentially, each trained independently, leading to potential cascading error propagation and complex parameter tuning challenges. In rethinking the mathematical modeling of the fusion and rectangling stages, we discovered that these processes can be effectively combined into a single, variety-intensity inpainting problem. Therefore, we propose the Simple and Robust Stitcher (SRStitcher), an efficient training-free image stitching method that …

arxiv cs.cv fusion image pipeline stitching type unified model