RECORD
Implicit Neural Image Stitching
Computer Vision and Pattern Recognition · Submitted 4 Sep 2023
abstract compiled · 2026-09-23 · no language-model review
The authors' main claim Our experiments show that our approach achieves improvement in resolving the low-definition imaging of the previous deep image stitching with favorable accelerated image-enhancing methods.
The compiled record
The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.
Open the compiled record
The abstract, sentence by sentence
Existing frameworks for image stitching often provide visually reasonable stitchings. However, they suffer from blurry artifacts and disparities in illumination, depth level, etc. Although the recent learning-based stitchings relax such disparities, the required methods impose sacrifice of image qualities failing to capture high-frequency details for stitched images. To address the problem, we propose a novel approach, implicit Neural Image Stitching (NIS) that extends arbitrary-scale super-resolution. Our method estimates Fourier coefficients of images for quality-enhancing warps. Then, the suggested model blends color mismatches and misalignment in the latent space and decodes the features into RGB values of stitched images. Our experiments show that our approach achieves improvement in resolving the low-definition imaging of the previous deep image stitching with favorable accelerated image-enhancing methods. Our source code is available at https://github.com/minshu-kim/NIS.
background contribution method result interpretation main claim
| # | sentence | role | the authors' own work | states the main conclusion | strength |
|---|---|---|---|---|---|
| s0 | Existing frameworks for image stitching often provide visually reasonable stitchings. | background 1.00 | 0.04 | 0.02 | asserted 0.94 |
| s1 | However, they suffer from blurry artifacts and disparities in illumination, depth level, etc. | background 0.78 | 0.06 | 0.02 | asserted 1.00 |
| s2 | Although the recent learning-based stitchings relax such disparities, the required methods impose sacrifice of image qualities failing to capture high-frequency details for stitched images. | gap 0.50 | 0.05 | 0.04 | asserted 1.00 |
| s3 | To address the problem, we propose a novel approach, implicit Neural Image Stitching (NIS) that extends arbitrary-scale super-resolution. | contribution 1.00 | 0.80 | 0.12 | asserted 0.98 |
| s4 | Our method estimates Fourier coefficients of images for quality-enhancing warps. | method 0.99 | 0.87 | 0.05 | asserted 1.00 |
| s5 | Then, the suggested model blends color mismatches and misalignment in the latent space and decodes the features into RGB values of stitched images. | method 1.00 | 0.82 | 0.05 | asserted 1.00 |
| s6 | Our experiments show that our approach achieves improvement in resolving the low-definition imaging of the previous deep image stitching with favorable accelerated image-enhancing methods. | result 1.00 | 0.95 | 0.74 | asserted 1.00 |
| s7 | Our source code is available at https://github.com/minshu-kim/NIS. | contribution 0.56 | 0.40 | 0.02 | asserted 0.94 |
What was examined, and what was not
- arXiv listing abstract v5
- full text
- figures and tables
- references and citing sentences
- relations to other papers
- P abstract
- S sections
- T body sentences
- C citations
- R cross-paper relations
The paper
Original abstract
Existing frameworks for image stitching often provide visually reasonable stitchings. However, they suffer from blurry artifacts and disparities in illumination, depth level, etc. Although the recent learning-based stitchings relax such disparities, the required methods impose sacrifice of image qualities failing to capture high-frequency details for stitched images. To address the problem, we propose a novel approach, implicit Neural Image Stitching (NIS) that extends arbitrary-scale super-resolution. Our method estimates Fourier coefficients of images for quality-enhancing warps. Then, the suggested model blends color mismatches and misalignment in the latent space and decodes the features into RGB values of stitched images. Our experiments show that our approach achieves improvement in resolving the low-definition imaging of the previous deep image stitching with favorable accelerated image-enhancing methods. Our source code is available at https://github.com/minshu-kim/NIS.
Human review
How the literature uses it
What this paper does with the arXiv papers it cites
What it cites · 5 sentences without a confident label
Uses the method or tool of 1 paper
-
We use Adam optimizer [ref] with $β_1=0.9$ and $β_2=0.999$.
Adam: A Method for Stochastic Optimization · Training Configurations · 1.00
Cites as background 4 papers
-
Image Stitching There are two branches under image stitching, view-fixed [ref] , and view-free tasks [ref] .
Video Stitching for Linear Camera Arrays · Related Work · 0.97
-
Image Stitching There are two branches under image stitching, view-fixed [ref] , and view-free tasks [ref] .
Learning Edge-Preserved Image Stitching from Large-Baseline Deep Homography · Related Work · 0.97
-
The first supervised and unsupervised deep homography estimators are proposed by Detone et al. and Nguyen et al. [ref] , respectively.
Deep Image Homography Estimation · Related Work · 0.96
Discussion
Know something about this paper? Leave a comment, or highlight any passage above to comment on it.