Unsupervised Deep Image Stitching: Reconstructing Stitched Features to Images

被引:127
|
作者
Nie, Lang [1 ,2 ]
Lin, Chunyu [1 ,2 ]
Liao, Kang [1 ,2 ]
Liu, Shuaicheng [3 ]
Zhao, Yao [1 ,2 ]
机构
[1] Beijing Jiaotong Univ, Inst Informat Sci, Beijing 100044, Peoples R China
[2] Beijing Key Lab Adv Informat Sci & Network Techno, Beijing 100044, Peoples R China
[3] Univ Elect Sci & Technol China, Sch Informat & Commun Engn, Chengdu 611731, Peoples R China
基金
中国国家自然科学基金;
关键词
Image stitching; Image reconstruction; Feature extraction; Image resolution; Estimation; Strain; Solid modeling; Computer vision; deep image stitching; deep homogrpahy estimation; HOMOGRAPHY;
D O I
10.1109/TIP.2021.3092828
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Traditional feature-based image stitching technologies rely heavily on feature detection quality, often failing to stitch images with few features or low resolution. The learning-based image stitching solutions are rarely studied due to the lack of labeled data, making the supervised methods unreliable. To address the above limitations, we propose an unsupervised deep image stitching framework consisting of two stages: unsupervised coarse image alignment and unsupervised image reconstruction. In the first stage, we design an ablation-based loss to constrain an unsupervised homography network, which is more suitable for large-baseline scenes. Moreover, a transformer layer is introduced to warp the input images in the stitching-domain space. In the second stage, motivated by the insight that the misalignments in pixel-level can be eliminated to a certain extent in feature-level, we design an unsupervised image reconstruction network to eliminate the artifacts from features to pixels. Specifically, the reconstruction network can be implemented by a low-resolution deformation branch and a high-resolution refined branch, learning the deformation rules of image stitching and enhancing the resolution simultaneously. To establish an evaluation benchmark and train the learning framework, a comprehensive real-world image dataset for unsupervised deep image stitching is presented and released. Extensive experiments well demonstrate the superiority of our method over other state-of-the-art solutions. Even compared with the supervised solutions, our image stitching quality is still preferred by users.
引用
收藏
页码:6184 / 6197
页数:14
相关论文
共 50 条
  • [21] Learning Pixel-wise Alignment for Unsupervised Image Stitching
    Jia, Qi
    Feng, Xiaomei
    Liu, Yu
    Fan, Xin
    Latecki, Longin Jan
    PROCEEDINGS OF THE 31ST ACM INTERNATIONAL CONFERENCE ON MULTIMEDIA, MM 2023, 2023, : 1392 - 1400
  • [22] An effective unsupervised image stitching method based on improved UDISNet
    Wang, Hongru
    Zhang, Jingtao
    Dai, Chaolei
    Cheng, Hu
    Wang, Jia
    ENGINEERING RESEARCH EXPRESS, 2025, 7 (01):
  • [23] Multimodal learning with only image data: A deep unsupervised model for street view image retrieval by fusing visual and scene text features of images
    Wu, Shangyou
    Yu, Wenhao
    Zhang, Yifan
    Huang, Mengqiu
    TRANSACTIONS IN GIS, 2024, 28 (03) : 486 - 508
  • [24] Unsupervised Learning of Deep-Learned Features from Breast Cancer Images
    Lee, Sanghoon
    Farley, Colton
    Shim, Simon
    Yoo, Wook-Sung
    Zhao, Yanjun
    Choi, Wookjin
    2020 IEEE 20TH INTERNATIONAL CONFERENCE ON BIOINFORMATICS AND BIOENGINEERING (BIBE 2020), 2020, : 740 - 745
  • [25] A Fast Unsupervised Image Stitching Model Based on Homography Estimation
    Ni, Jianjun
    Li, Yingqi
    Ke, Chunyan
    Zhang, Ziru
    Cao, Weidong
    Yang, Simon X.
    IEEE SENSORS JOURNAL, 2024, 24 (18) : 29452 - 29467
  • [26] Image Stitching Based on Local Symmetry Features
    Yang Di
    Bo Yu-ming
    Zhao Gao-peng
    2014 33RD CHINESE CONTROL CONFERENCE (CCC), 2014, : 4641 - 4646
  • [27] Deep Feature Extraction for Panoramic Image Stitching
    Van-Dung Hoang
    Diem-Phuc Tran
    Nguyen Gia Nhu
    The-Anh Pham
    Van-Huy Pham
    INTELLIGENT INFORMATION AND DATABASE SYSTEMS (ACIIDS 2020), PT II, 2020, 12034 : 141 - 151
  • [28] Image stitching via deep homography estimation
    Zhao, Qiang
    Ma, Yike
    Zhu, Chen
    Yao, Chunfeng
    Feng, Bailan
    Dai, Feng
    NEUROCOMPUTING, 2021, 450 : 219 - 229
  • [29] Deep Rectangling for Image Stitching: A Learning Baseline
    Nie, Lang
    Lin, Chunyu
    Liao, Kang
    Liu, Shuaicheng
    Zhao, Yao
    2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2022), 2022, : 5730 - 5738
  • [30] Reflection hologram for reconstructing deep images
    Taketomi, Y
    Kubota, T
    OPTICS LETTERS, 1997, 22 (22) : 1725 - 1727