SOLAR: Second-Order Loss and Attention for Image Retrieval

Ng, Tony; Balntas, Vassileios; Tian, Yanling; Mikolajczyk, Krystian

doi:10.1007/978-3-030-58595-2_16

Cited by 82 publications

(47 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [94] a Quadratic Hinge Triplet loss which constraints first-order similarity is paired with a loss that regularizes second-order similarity for local descriptor learning. This idea is also applied to global descriptors learning in [95]. In [96], the triplet loss is modified by setting the distance to the hardest negative example to a constant value, so that the corresponding derivative of the loss is set to zero.…”

Section: B Learning To Rankmentioning

confidence: 99%

“…Another solution to produce more informative attention maps is to leverage second-order spatial information, as done in [95] using a non-local block [171]. Second order spatial information allows to generate a feature map in which local features reflect the correlations between all spatial locations, in contrast to first order features where each local feature has a limited receptive field.…”

Section: ) Attention Modules and Weighting Masksmentioning

confidence: 99%

See 1 more Smart Citation

A Survey on Deep Visual Place Recognition

2021

View full text Add to dashboard Cite

In recent years visual place recognition (VPR), i.e., the problem of recognizing the location of images, has received considerable attention from multiple research communities, spanning from computer vision to robotics and even machine learning. This interest is fueled on one hand by the relevance that visual place recognition holds for many applications and on the other hand by the unsolved challenge of making these methods perform reliably in different conditions and environments. This paper presents a survey of the state-of-the-art of research on visual place recognition, focusing on how it has been shaped by the recent advances in deep learning. We start discussing the image representations used in this task and how they have evolved from using hand-crafted to deep-learned features. We further review how metric learning techniques are used to get more discriminative representations, as well as techniques for dealing with occlusions, distractors, and shifts in the visual domain of the images. The survey also provides an overview of the specific solutions that have been proposed for applications in robotics and with aerial imagery. Finally the survey provides a summary of datasets that are used in visual place recognition, highlighting their different characteristics. INDEX TERMS Visual place recognition, image representation learning, deep learning.

show abstract

Section: B Learning To Rankmentioning

confidence: 99%

Section: ) Attention Modules and Weighting Masksmentioning

confidence: 99%

A Survey on Deep Visual Place Recognition

2021

View full text Add to dashboard Cite

show abstract

“…See Figure 5 for the principle of deep descriptors. Global descriptors vary depending on how they process the tensor of activations: simple pooling operations such as sum [19], max [20], or generalized mean (GeM) [21]; or more complex operations such as cross-dimensional weighting [22] or second-order attention maps [23]. Local descriptors rely on a selection operation: the tensor of activations is considered as a set of local descriptors, which must be filtered with attention mechanisms to only keep the most discriminative ones.…”

Section: Retrieval Frameworkmentioning

confidence: 99%

Connecting Images through Sources: Exploring Low-Data, Heterogeneous Instance Retrieval

2021

View full text Add to dashboard Cite

Along with a new volume of images containing valuable information about our past, the digitization of historical territorial imagery has brought the challenge of understanding and interconnecting collections with unique or rare representation characteristics, and sparse metadata. Content-based image retrieval offers a promising solution in this context, by building links in the data without relying on human supervision. However, while the latest propositions in deep learning have shown impressive results in applications linked to feature learning, they often rely on the hypothesis that there exists a training dataset matching the use case. Increasing generalization and robustness to variations remains an open challenge, poorly understood in the context of real-world applications. Introducing the alegoria benchmark, containing multi-date vertical and oblique aerial digitized photography mixed with more modern street-level pictures, we formulate the problem of low-data, heterogeneous image retrieval, and propose associated evaluation setups and measures. We propose a review of ideas and methods to tackle this problem, extensively compare state-of-the-art descriptors and propose a new multi-descriptor diffusion method to exploit their comparative strengths. Our experiments highlight the benefits of combining descriptors and the compromise between absolute and cross-domain performance.

show abstract

“…Moreover, CNN can give a global descriptor of an image such as LBP [85]. The proposed works [70,69,68,67], transform an input image into a global representation. Descriptors based on deep learning have been shown to be more robust against rotation and illumination changes than classical descriptors.…”

Section: Fig 2 Bag Of Visual Words Modelmentioning

confidence: 99%

Leveraging semantic segmentation for hybrid image retrieval methods

Ouni

Royer

Chevaldonné

et al. 2021

Neural Comput & Applic

View full text Add to dashboard Cite

Content Based Image Retrieval (CBIR) is the task of finding images in a database that are the most similar to the input query based on its visual characteristics. Several methods from the state of the art based on visual methods (Bag of visual words, VLAD, ...) or recent deep leaning methods try to solve the CBIR problem. In particular, Deep learning is a new field and used for several vision applications including CBIR. But, even with the increase of the performance of deep learning algorithms, this problem is still a challenge in computer vision. In this work, we propose three different methodologies combining deep learning based semantic segmentation and visual features. We show experimentally that by exploiting semantic information in the CBIR context leads to an increase in the retrieval accuracy. We study the performance of the proposed approach on eight different datasets (Wang, Corel-10k, Corel-5k, GHIM-10K, MSRC V1, MSRC V2, Linnaeus, NUS-WIDE)

show abstract

SOLAR: Second-Order Loss and Attention for Image Retrieval

Cited by 82 publications

References 48 publications

A Survey on Deep Visual Place Recognition

A Survey on Deep Visual Place Recognition

Connecting Images through Sources: Exploring Low-Data, Heterogeneous Instance Retrieval

Leveraging semantic segmentation for hybrid image retrieval methods

Contact Info

Product

Resources

About