Abstract:Image stitching aims at stitching the images taken from different viewpoints into an image with a wider field of view. Existing methods warp the target image to the reference image using the estimated warp function, and a homography is one of the most commonly used warping functions. However, when images have large parallax due to non-planar scenes and translational motion of a camera, the homography cannot fully describe the mapping between two images. Existing approaches based on global or local homography estimation are not free from this problem and suffer from undesired artifacts due to parallax. In this paper, instead of relying on the homography-based warp, we propose a novel deep image stitching framework exploiting the pixel-wise warp field to handle the large-parallax problem. The proposed deep image stitching framework consists of two modules: Pixel-wise Warping Module (PWM) and Stitched Image Generating Module (SIGMo). PWM employs an optical flow estimation model to obtain pixel-wise warp of the whole image, and relocates the pixels of the target image with the obtained warp field. SIGMo blends the warped target image and the reference image while eliminating unwanted artifacts such as misalignments, seams, and holes that harm the plausibility of the stitched result. For training and evaluating the proposed framework, we build a large-scale dataset that includes image pairs with corresponding pixel-wise ground truth warp and sample stitched result images. We show that the results of the proposed framework are qualitatively superior to those of the conventional methods, especially when the images have large parallax. The code and the proposed dataset will be publicly available soon.

A content-aware image stitching algorithm for mobile multimedia devices

Remote Sensing Image Stitch Using Modified Structure Deformation

A Robust Image Stitching Based on Solution Set Smoothing

Line-preserving video stitching for asymmetric cameras

Research on Improved Multi-Channel Image Stitching Technology Based on Fast Algorithms

A new optimal seam selection method for airborne image stitching

Real-Time Multi-Camera Video Stitching Algorithm with Adaptive Sensing of Illumination and Dynamic Objects

Fast Color Blending for Seamless Image Stitching.

SC-AOF: A Sliding Camera and Asymmetric Optical-Flow-Based Blending Method for Image Stitching

A Simple Yet Effective Image Stitching with Computational Suture Zone.

Seamless Video Stitching from Hand-held Camera Inputs.

Semantic segmentation guided feature point classification and seam fusion for image stitching

Pixel-wise Deep Image Stitching

Image Stitching Algorithm Based on Area Blocking and Optimal Seam

A Seamless Image-Stitching Method Based on Human Visual Discrimination and Attention

Real-Time Image Mosaic based on Optimal Seam and Multiband Blend

A Novel Hybrid Color Registration Algorithm for Image Stitching

Content-Seam-Preserving Multi-Alignment Network for Visual-Sensor-Based Image Stitching

Object-centered image stitching

A Color Correction Method Based on Incremental Multilevel Iterative Histogram Matching

Paper Stitching Using Maximum Tolerant Seam under Local Distortions.