Summarizing visual data using bidirectional similarity

By Denis Simakov, Yaron Caspi, Eli Shechtman, Michal Irani. 2008

674 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

Regionwise Generative Adversarial Image Inpainting for Large Missing Areas (2019, background): “It has been widely applied in many scenarios [1]–[3], such as image recovery (removing photograph scratches and text occlusions), photograph editing (removing unwanted objects, face editing), image encoding and transmission (loss of blocky image content caused by network packet loss during image tra”
Free-Form Image Inpainting With Gated Convolution (2018, methodology): “Further, Simakov et al . propose bidirectional similarity synthesis approach [Simakov et al. 2008] to better capture and summarize non-stationary visual data.”
Generative Image Inpainting with Contextual Attention (2018, methodology): “[34] propose a bidirectional patch similarity-based scheme to better model nonstationary visual data for re-targeting and inpainting applications.”
Image Inpainting for Irregular Holes Using Partial Convolutions (2018, background): “However, these steps often come at a large computation cost such as in [26].”
Discovering important people and objects for egocentric video summarization (2012, background): “A retargeting approach aims to simultaneously preserve an orig inal video’s content while reducing artifacts [24], but unl ike our approach, does not attempt to characterize the varying degrees of object importance.”
BlobGAN: Spatially Disentangled Scene Representations (2022, background): “This interaction is related to traditional “image reshuffling” where rearrangement of image content is done in pixel space [81, 4, 70].”
The patch transform and its applications to image editing (2008, methodology): “Related functionalities, obtained using a different approach, are described in [ 25].”
PEPSI : Fast Image Inpainting With Parallel Decoding Network (2019, methodology): “In contrast to the diffusion-based method, the patch-based method samples patches from the background region and then pastes them into the hole region [22, 28].”
Regenerative morphing (2010, methodology): “Motion constraints The original BDS method [16] allowed patches to search their nearest neighbor at any location in the second image.”
DHSNet: Deep Hierarchical Saliency Network for Salient Object Detection (2016, background): “In recent years, researchers have developed many computational models for salient object detection and applied them to benefit many other applications, such as image summarization [1], segmentation [2], retrieval [3], and editing [4].”
The Generalized PatchMatch Correspondence Algorithm (2010, background): “Moreover, many analysis applications [8–11] and synthesis applications [12–15] inherently require dense global correspondences for adequate performance.”
High Dynamic Range Imaging (2015, methodology): “The patch match algorithm, which exploits self-similarities in images [12, 141], is used in this application to optimize local similarity of the reconstructed HDR image to all input exposures.”
Coarse-to-Fine Image Inpainting via Region-wise Convolutions and Non-Local Correlation (2019, methodology): “It can be widely applied in many tasks [Barnes et al., 2009a; Newson et al., 2014; Park et al., 2017; Simakov et al., 2008], such as photo editing, image-based rendering, computational photography, etc. Till now, there have been many methods proposed for generating desirable ∗Corresponding…”
Deep Saliency with Encoded Low Level Distance Map and High Level Features (2016, background): “s received a great deal of attention in computer vision and cognitive science because of its wide range of applications such as content-aware image cropping [22] and resizing [3], video summarization [24], object detection [20], and person re-identiﬁcation [31]. Various papers such as DRFI [13], GMR”
Globally and locally consistent image completion (2017, methodology): “Although many approaches have been proposed for image completion, such as patch-based image synthesis [Barnes et al. 2009; Darabi et al. 2012; Huang et al. 2014; Simakov et al. 2008; Wexler et al. 2007], it remains a challenging problem because it often requires high-level recognition of scenes.”
Scale and object aware image retargeting for thumbnail browsing (2011, methodology): “As pointed out in previous sections and also observed in [20], many discrete retargeting methods generate excellent results in general but they sometimes create serious artifacts when the target has a size much smaller than the input.”
Image Fine-grained Inpainting (2020, positive): “[22] proposed bidirectional similarity scheme to capture better and summarize non-stationary visual data.”