Image Location Inference by Multisaliency Enhancement

Qian, Xueming; Wang, Huan; Zhao, Yisi; Hou, Xingsong; Hong, Richang; Wang, Meng; Tang, Yuan Yan

doi:10.1109/tmm.2016.2638207

Cited by 30 publications

(8 citation statements)

References 69 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A. Datasets VQ approaches are so general that can be applied to different kinds of features, including features for image [43], [44], video [45], text [46], [47], or sensing data [48], [49]. To better compare our framework with previous VQ approaches, we mainly focus on image features in the experiment below.…”

Section: Experiments and Discussionmentioning

confidence: 99%

Robust Quantization for General Similarity Search

Guo

Ding

Han

2018

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

The recent years have witnessed the emerging of vector quantization (VQ) techniques for efficient similarity search. VQ partitions the feature space into a set of codewords and encodes data points as integer indices using the codewords. Then the distance between data points can be efficiently approximated by simple memory lookup operations. By the compact quantization, the storage cost, and searching complexity are significantly reduced, thereby facilitating efficient large-scale similarity search. However, the performance of several celebrated VQ approaches degrades significantly when dealing with noisy data. In addition, it can barely facilitate a wide range of applications as the distortion measurement only limits to ℓ norm. To address the shortcomings of the squared Euclidean (ℓ norm) loss function employed by the VQ approaches, in this paper, we propose a novel robust and general VQ framework, named RGVQ, to enhance both robustness and generalization of VQ approaches. Specifically, a ℓ-norm loss function is proposed to conduct the ℓ-norm similarity search, rather than the ℓ norm search, and the q-th order loss is used to enhance the robustness. Despite the fact that changing the loss function to ℓ norm makes VQ approaches more robust and generic, it brings us a challenge that a non-smooth and non-convex orthogonality constrained ℓ-norm function has to be minimized. To solve this problem, we propose a novel and efficient optimization scheme and specify it to VQ approaches and theoretically prove its convergence. Extensive experiments on benchmark data sets demonstrate that the proposed RGVQ is better than the original VQ for several approaches, especially when searching similarity in noisy data.

show abstract

Section: Experiments and Discussionmentioning

confidence: 99%

Robust Quantization for General Similarity Search

Guo

Ding

Han

2018

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

show abstract

“…Early works of image-based localization can be divided into two main categories: retrieval based approach and 3D model-based approach (or direct search approach). Retrieval based methods [13], [25], [14], [12], [26] are closely related to image retrieval by matching query features against geo-tagged database images. This matching will result in a set of similar images according to the query.…”

Section: A Image Based Localizationmentioning

confidence: 99%

On-Device Scalable Image-Based Localization via Prioritized Cascade Search and Fast One-Many RANSAC

Tran¹,

Tan²,

Doan³

et al. 2019

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

We present the design of an entire on-device system for large-scale urban localization using images. The proposed design integrates compact image retrieval and 2D-3D correspondence search to estimate the location in extensive city regions. Our design is GPS agnostic and does not require network connection. In order to overcome the resource constraints of mobile devices, we propose a system design that leverages the scalability advantage of image retrieval and accuracy of 3D model-based localization. Furthermore, we propose a new hashing-based cascade search for fast computation of 2D-3D correspondences. In addition, we propose a new one-many RANSAC for accurate pose estimation. The new one-many RANSAC addresses the challenge of repetitive building structures (e.g. windows, balconies) in urban localization. Extensive experiments demonstrate that our 2D-3D correspondence search achieves state-of-the-art localization accuracy on multiple benchmark datasets. Furthermore, our experiments on a large Google Street View (GSV) image dataset show the potential of large-scale localization entirely on a typical mobile device.Index Terms-Image-based localization, on-device localization, image retrieval, 2D-3D correspondence search, hashing, RANSAC

show abstract

“…Feature co-occurrence is often adopted in computer vision [23,39,53] and visual search [49,51,54]. The idea is also successfully utilized in recommender systems [55].…”

Section: Inter-modal Modelingmentioning

confidence: 99%

Modeling Dyadic and Group Impressions with Intermodal and Interperson Features

Okada

Nguyen

Aran

et al. 2019

ACM Trans. Multimedia Comput. Commun. Appl.

View full text Add to dashboard Cite

This paper proposes a novel feature-extraction framework for inferring impressed personality traits, emergent leadership skills, communicative competence and hiring decisions. The proposed framework extracts multimodal features, describing each participant's nonverbal activities. It captures intermodal and inter-person relationships in interactions and captures how the target interactor generates nonverbal behavior when other interactors also generate nonverbal behavior. The intermodal and inter-person patterns are identified as frequent co-occurring events based on clustering from multimodal sequences. The proposed framework is applied to the SONVB corpus, which is an audiovisual dataset collected from dyadic job interviews, and the ELEA audiovisual data corpus, which is a dataset collected from group meetings. We evaluate the framework on a binary classification task involving 15 impression variables from the two data corpora. The experimental results show that the model trained with co-occurrence features is more accurate than previous models for 14 out of 15 traits. CCS Concepts: • Human-centered computing → Social engineering (social sciences); • Computing methodologies → Discourse, dialogue and pragmatics; • Information systems → Clustering;

show abstract

Image Location Inference by Multisaliency Enhancement

Cited by 30 publications

References 69 publications

Robust Quantization for General Similarity Search

Robust Quantization for General Similarity Search

On-Device Scalable Image-Based Localization via Prioritized Cascade Search and Fast One-Many RANSAC

Modeling Dyadic and Group Impressions with Intermodal and Interperson Features

Contact Info

Product

Resources

About