scholarly journals Fine-Grained Visual Textual Alignment for Cross-Modal Retrieval Using Transformer Encoders

2021 ◽  
Vol 17 (4) ◽  
pp. 1-23
Author(s):  
Nicola Messina ◽  
Giuseppe Amato ◽  
Andrea Esuli ◽  
Fabrizio Falchi ◽  
Claudio Gennaro ◽  
...  
Keyword(s):  
Information Retrieval ◽  
Large Scale ◽  
State Of The Art ◽  
Retrieval Task ◽  
Fine Grained ◽  
Retrieval Systems ◽  
Sentence Retrieval ◽  
Novel Approach ◽  
Sentence Level ◽  
Sentence Matching

Despite the evolution of deep-learning-based visual-textual processing systems, precise multi-modal matching remains a challenging task. In this work, we tackle the task of cross-modal retrieval through image-sentence matching based on word-region alignments, using supervision only at the global image-sentence level. Specifically, we present a novel approach called Transformer Encoder Reasoning and Alignment Network (TERAN). TERAN enforces a fine-grained match between the underlying components of images and sentences (i.e., image regions and words, respectively) to preserve the informative richness of both modalities. TERAN obtains state-of-the-art results on the image retrieval task on both MS-COCO and Flickr30k datasets. Moreover, on MS-COCO, it also outperforms current approaches on the sentence retrieval task. Focusing on scalable cross-modal information retrieval, TERAN is designed to keep the visual and textual data pipelines well separated. Cross-attention links invalidate any chance to separately extract visual and textual features needed for the online search and the offline indexing steps in large-scale retrieval systems. In this respect, TERAN merges the information from the two domains only during the final alignment phase, immediately before the loss computation. We argue that the fine-grained alignments produced by TERAN pave the way toward the research for effective and efficient methods for large-scale cross-modal information retrieval. We compare the effectiveness of our approach against relevant state-of-the-art methods. On the MS-COCO 1K test set, we obtain an improvement of 5.7% and 3.5% respectively on the image and the sentence retrieval tasks on the Recall@1 metric. The code used for the experiments is publicly available on GitHub at https://github.com/mesnico/TERAN .

scholarly journals ShadingNet: Image Intrinsics by Fine-Grained Shading Decomposition

2021 ◽  
Author(s):  
Anil S. Baslamisli ◽  
Partha Das ◽  
Hoang-An Le ◽  
Sezer Karaoglu ◽  
Theo Gevers
Keyword(s):  
Neural Network ◽  
Large Scale ◽  
State Of The Art ◽  
Image Decomposition ◽  
Natural Environments ◽  
Decomposition Algorithms ◽  
Ambient Light ◽  
Fine Grained ◽  
Large Scale Dataset ◽  
Direct Illumination

AbstractIn general, intrinsic image decomposition algorithms interpret shading as one unified component including all photometric effects. As shading transitions are generally smoother than reflectance (albedo) changes, these methods may fail in distinguishing strong photometric effects from reflectance variations. Therefore, in this paper, we propose to decompose the shading component into direct (illumination) and indirect shading (ambient light and shadows) subcomponents. The aim is to distinguish strong photometric effects from reflectance variations. An end-to-end deep convolutional neural network (ShadingNet) is proposed that operates in a fine-to-coarse manner with a specialized fusion and refinement unit exploiting the fine-grained shading model. It is designed to learn specific reflectance cues separated from specific photometric effects to analyze the disentanglement capability. A large-scale dataset of scene-level synthetic images of outdoor natural environments is provided with fine-grained intrinsic image ground-truths. Large scale experiments show that our approach using fine-grained shading decompositions outperforms state-of-the-art algorithms utilizing unified shading on NED, MPI Sintel, GTA V, IIW, MIT Intrinsic Images, 3DRMS and SRD datasets.

scholarly journals Multi-Task Driven Feature Models for Thermal Infrared Tracking

2020 ◽  
Vol 34 (07) ◽  
pp. 11604-11611 ◽  
Author(s):  
Qiao Liu ◽  
Xin Li ◽  
Zhenyu He ◽  
Nana Fan ◽  
Di Yuan ◽  
...  
Keyword(s):  
Large Scale ◽  
State Of The Art ◽  
Thermal Infrared ◽  
Training Dataset ◽  
Feature Models ◽  
Relative Gain ◽  
Tir Domain ◽  
Fine Grained ◽  
Rgb Images ◽  
Infrared Tracking

Existing deep Thermal InfraRed (TIR) trackers usually use the feature models of RGB trackers for representation. However, these feature models learned on RGB images are neither effective in representing TIR objects nor taking fine-grained TIR information into consideration. To this end, we develop a multi-task framework to learn the TIR-specific discriminative features and fine-grained correlation features for TIR tracking. Specifically, we first use an auxiliary classification network to guide the generation of TIR-specific discriminative features for distinguishing the TIR objects belonging to different classes. Second, we design a fine-grained aware module to capture more subtle information for distinguishing the TIR objects belonging to the same class. These two kinds of features complement each other and recognize TIR objects in the levels of inter-class and intra-class respectively. These two feature models are learned using a multi-task matching framework and are jointly optimized on the TIR tracking task. In addition, we develop a large-scale TIR training dataset to train the network for adapting the model to the TIR domain. Extensive experimental results on three benchmarks show that the proposed algorithm achieves a relative gain of 10% over the baseline and performs favorably against the state-of-the-art methods. Codes and the proposed TIR dataset are available at https://github.com/QiaoLiuHit/MMNet.

scholarly journals Predicting and Analyzing Language Specificity in Social Media Posts

2019 ◽  
Vol 33 ◽  
pp. 6415-6422 ◽  
Author(s):  
Yifan Gao ◽  
Yang Zhong ◽  
Daniel Preoţiuc-Pietro ◽  
Junyi Jessy Li
Keyword(s):  
Social Media ◽  
Computational Linguistics ◽  
Large Scale ◽  
Pearson Correlation ◽  
Absolute Error ◽  
Fine Grained ◽  
Argumentation Mining ◽  
Sentence Level ◽  
Prediction Systems ◽  
Mental Health Factors

In computational linguistics, specificity quantifies how much detail is engaged in text. It is an important characteristic of speaker intention and language style, and is useful in NLP applications such as summarization and argumentation mining. Yet to date, expert-annotated data for sentence-level specificity are scarce and confined to the news genre. In addition, systems that predict sentence specificity are classifiers trained to produce binary labels (general or specific).We collect a dataset of over 7,000 tweets annotated with specificity on a fine-grained scale. Using this dataset, we train a supervised regression model that accurately estimates specificity in social media posts, reaching a mean absolute error of 0.3578 (for ratings on a scale of 1-5) and 0.73 Pearson correlation, significantly improving over baselines and previous sentence specificity prediction systems. We also present the first large-scale study revealing the social, temporal and mental health factors underlying language specificity on social media.

IR Research and Innovation in Commercial Online Systems: An Exploratory Survey

2013 ◽  
Author(s):  
Sherry Koshman ◽  
Edie Rasmussen
Keyword(s):  
Information Retrieval ◽  
Large Scale ◽  
Interactive Systems ◽  
Information Industry ◽  
Online Systems ◽  
Research And Innovation ◽  
Retrieval Systems ◽  
Information Retrieval Systems ◽  
Operational Systems ◽  
Mcgill University

From the 1994 CAIS Conference: The Information Industry in Transition McGill University, Montreal, Quebec. May 25 - 27, 1994."Conventional" information retrieval systems (IRS), originating in the research of the 11950s and 1960s, are based on keyword matching and the application of Boolean operators to produce a set of retrieved documents from a database. In the ensuing years, research in information retrieval has identified a number of innovations (for example, automatic weighting of terms, ranked output, and relevance feedback) which have the potential to significantly enhance the performance of IRS, though commercial vendors have been slow to incorporate these changes into their systems. This was the situation in 1988 which led Radecki, in a special issue of Information Processing & Management, to examine the potential for improvements in conventional Boolean retrieval systems, and explore the reasons why these improvements had not been implemented in operational systems. Over the last five years, this position has begun to change as commercial vendors such as Dialog, Dow Jones, West Publishing, and Mead have implemented new, non-Boolean features in their systems, including natural language input, weighted keyword terms, and document ranking. This paper identifies some of the significant findings of IR research and compares them to the implementation of non-Boolean features in such systems. The preliminary survey of new features in commercial systems suggests the need for new methods of evaluation, including the development of evalutation measures appropriate to large-scale, interactive systems.

scholarly journals Large-Scale Multi-View Subspace Clustering in Linear Time

2020 ◽  
Vol 34 (04) ◽  
pp. 4412-4419 ◽  
Author(s):  
Zhao Kang ◽  
Wangtao Zhou ◽  
Zhitong Zhao ◽  
Junming Shao ◽  
Meng Han ◽  
...  
Keyword(s):  
Large Scale ◽  
State Of The Art ◽  
Linear Time ◽  
Subspace Clustering ◽  
Data Sets ◽  
Clustering Methods ◽  
Single View ◽  
Novel Approach ◽  
Points Of View ◽  
Effectiveness And Efficiency

A plethora of multi-view subspace clustering (MVSC) methods have been proposed over the past few years. Researchers manage to boost clustering accuracy from different points of view. However, many state-of-the-art MVSC algorithms, typically have a quadratic or even cubic complexity, are inefficient and inherently difficult to apply at large scales. In the era of big data, the computational issue becomes critical. To fill this gap, we propose a large-scale MVSC (LMVSC) algorithm with linear order complexity. Inspired by the idea of anchor graph, we first learn a smaller graph for each view. Then, a novel approach is designed to integrate those graphs so that we can implement spectral clustering on a smaller graph. Interestingly, it turns out that our model also applies to single-view scenario. Extensive experiments on various large-scale benchmark data sets validate the effectiveness and efficiency of our approach with respect to state-of-the-art clustering methods.

scholarly journals SafeNet: Scale-normalization and Anchor-based Feature Extraction Network for Person Re-identification

2018 ◽  
Author(s):  
Kun Yuan ◽  
Qian Zhang ◽  
Chang Huang ◽  
Shiming Xiang ◽  
Chunhong Pan
Keyword(s):  
Large Scale ◽  
State Of The Art ◽  
Spatial Distributions ◽  
Body Parts ◽  
Retrieval Task ◽  
Multi Scale ◽  
Aspect Ratios ◽  
Scale Normalization ◽  
Full Body

Person Re-identification (ReID) is a challenging retrieval task that requires matching a person's image across non-overlapping camera views. The quality of fulfilling this task is largely determined on the robustness of the features that are used to describe the person. In this paper, we show the advantage of jointly utilizing multi-scale abstract information to learn powerful features over full body and parts. A scale normalization module is proposed to balance different scales through residual-based integration. To exploit the information hidden in non-rigid body parts, we propose an anchor-based method to capture the local contents by stacking convolutions of kernels with various aspect ratios, which focus on different spatial distributions. Finally, a well-defined framework is constructed for simultaneously learning the representations of both full body and parts. Extensive experiments conducted on current challenging large-scale person ReID datasets, including Market1501, CUHK03 and DukeMTMC, demonstrate that our proposed method achieves the state-of-the-art results.

scholarly journals Large-scale image search with text for information retrieval

2021 ◽  
Vol 4 (1) ◽  
pp. 87-89
Author(s):  
Janardan Bhatta
Keyword(s):  
Information Retrieval ◽  
Language Processing ◽  
Large Scale ◽  
Image Feature ◽  
Image Search ◽  
Search Results ◽  
Retrieval Systems ◽  
Information Retrieval Systems ◽  
Text Features ◽  
Text Query

Searching images in a large database is a major requirement in Information Retrieval Systems. Expecting image search results based on a text query is a challenging task. In this paper, we leverage the power of Computer Vision and Natural Language Processing in Distributed Machines to lower the latency of search results. Image pixel features are computed based on contrastive loss function for image search. Text features are computed based on the Attention Mechanism for text search. These features are aligned together preserving the information in each text and image feature. Previously, the approach was tested only in multilingual models. However, we have tested it in image-text dataset and it enabled us to search in any form of text or images with high accuracy.

scholarly journals Dig into Multi-modal Cues for Video Retrieval with Hierarchical Alignment

2021 ◽  
Author(s):  
Wenzhe Wang ◽  
Mengdan Zhang ◽  
Runnan Chen ◽  
Guanyu Cai ◽  
Penghao Zhou ◽  
...  
Keyword(s):  
Video Retrieval ◽  
State Of The Art ◽  
Text Retrieval ◽  
Local Alignment ◽  
Global Alignment ◽  
Retrieval Task ◽  
Fine Grained ◽  
Global View ◽  
High Level ◽  
Semantic Correspondence

Multi-modal cues presented in videos are usually beneficial for the challenging video-text retrieval task on internet-scale datasets. Recent video retrieval methods take advantage of multi-modal cues by aggregating them to holistic high-level semantics for matching with text representations in a global view. In contrast to this global alignment, the local alignment of detailed semantics encoded within both multi-modal cues and distinct phrases is still not well conducted. Thus, in this paper, we leverage the hierarchical video-text alignment to fully explore the detailed diverse characteristics in multi-modal cues for fine-grained alignment with local semantics from phrases, as well as to capture a high-level semantic correspondence. Specifically, multi-step attention is learned for progressively comprehensive local alignment and a holistic transformer is utilized to summarize multi-modal cues for global alignment. With hierarchical alignment, our model outperforms state-of-the-art methods on three public video retrieval datasets.

Large-Scale Multi-modal Distance Metric Learning with Application to Content-Based Information Retrieval and Image Classification

2020 ◽  
Vol 34 (13) ◽  
pp. 2050034
Author(s):  
Ali Salim Rasheed ◽  
Davood Zabihzadeh ◽  
Sumia Abdulhussien Razooqi Al-Obaidi
Keyword(s):  
Information Retrieval ◽  
Image Classification ◽  
Large Scale ◽  
Learning Algorithm ◽  
State Of The Art ◽  
Metric Learning ◽  
Random Projection ◽  
Linear Projection ◽  
Modal Data ◽  
Related Data

Metric learning algorithms aim to make the conceptually related data items closer and keep dissimilar ones at a distance. The most common approach for metric learning on the Mahalanobis method. Despite its success, this method is limited to find a linear projection and also suffer from scalability respecting both the dimensionality and the size of input data. To address these problems, this paper presents a new scalable metric learning algorithm for multi-modal data. Our method learns an optimal metric for any feature set of the multi-modal data in an online fashion. We also combine the learned metrics with a novel Passive/Aggressive (PA)-based algorithm which results in a higher convergence rate compared to the state-of-the-art methods. To address scalability with respect to dimensionality, Dual Random Projection (DRP) is adopted in this paper. The present method is evaluated on some challenging machine vision datasets for image classification and Content-Based Information Retrieval (CBIR) tasks. The experimental results confirm that the proposed method significantly surpasses other state-of-the-art metric learning methods in most of these datasets in terms of both accuracy and efficiency.

Sign in / Sign up

Export Citation Format

Share Document