Large-Scale Face Image Retrieval System at Attribute Level Based on Facial Attribute Ontology and Deep Neuron Network

Nguyen, Hung Manh; Ly, Ngoc Q.; Phung, Trang T. T.

doi:10.1007/978-3-319-75420-8_51

Cited by 12 publications

(12 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thirdly, Multitask Learning (MTL) has been proved to be effective in AL, such as MCNN-AUX (Hand et al 2016), MT-RBM PCA (Ehrlich et al, 2016), DMTL (Han et al, 2017), MOON (Rudd et al, 2016), and FAO (Nguyen et al, 2018).…”

Section: Figure 5 Attribute Learning Approachesmentioning

confidence: 99%

See 1 more Smart Citation

An Overview of Facial Attribute Learning

Trang¹,

Fukuzawa²,

Ngoc³

2021

Tạp chí Khoa học

View full text Add to dashboard Cite

Thuộc tính mặt người là thông tin hữu ích cho việc xây dựng các ứng dụng như nhận dạng, tìm kiếm và giám sát khuôn mặt người. Do đó, chúng rất quan trọng đối với các nhiệm vụ phân tích khuôn mặt khác nhau. Nhiều thuật toán học thuộc tính khuôn mặt người đã và đang được phát triển để tự động phát hiện các thuộc tính trong nhiều năm qua. Trong bài báo này, chúng tôi khảo sát một số phương pháp điển hình về học thuộc tính khuôn mặt người. Chúng tôi chia ra năm loại chính của các phương pháp: (1) Học truyền thống, (2) Học sâu đơn nhiệm, (3) Học sâu đa nhiệm, (4) Giải quyết vấn đề mất cân bằng dữ liệu và (5) Thuộc tính khuôn mặt dựa vào phả hệ tri thức. Các phương pháp bao gồm từ học truyền thống đến học sâu, cùng với các phương pháp hỗ trợ giải quyết bài toán lỗ hổng ngữ nghĩa dựa trên phả hệ tri thức và giải quyết sự mất cân bằng dữ liệu. Đối với mỗi phương pháp trong mỗi loại, chúng tôi thảo luận về các lí thuyết cơ bản cũng như điểm mạnh, điểm yếu và sự khác biệt của chúng. Chúng tôi cũng so sánh hiệu suất của chúng trên bộ dữ liệu tiêu chuẩn. Cuối cùng, dựa trên đặc điểm và đóng góp của các phương pháp, chúng tôi đưa ra kết luận và hướng nghiên cứu trong tương lai để giải quyết vấn đề học thuộc tính khuôn mặt. bài khảo sát này sẽ giúp các nhà nghiên cứu có góc nhìn tổng quan nhanh để xây dựng các ứng dụng khuôn mặt người trong tương lai cũng như các nghiên cứu mới.

show abstract

Section: Figure 5 Attribute Learning Approachesmentioning

confidence: 99%

“…Finally, using ontology in AL to support semantic of images is also common such as in a study by Nguyen et al (2018), Ly et al (2019), Ly et al (2020).…”

Section: Figure 5 Attribute Learning Approachesmentioning

confidence: 99%

An Overview of Facial Attribute Learning

Trang¹,

Fukuzawa²,

Ngoc³

2021

Tạp chí Khoa học

View full text Add to dashboard Cite

show abstract

“…These types of systems find images within a database that are similar to query. Conilione et al [14], Nguyen et al [15], and Wang et al [16] attempted to discover semantic labels for facial features for use in such systems.…”

Section: Face Attribute Prediction and Annotationmentioning

confidence: 99%

A Method for Human Facial Image Annotation on Low Power Consumption Autonomous Devices

Hachaj

2020

Sensors

View full text Add to dashboard Cite

This paper proposes a classifier designed for human facial feature annotation, which is capable of running on relatively cheap, low power consumption autonomous microcomputer systems. An autonomous system is one that depends only on locally available hardware and software—for example, it does not use remote services available through the Internet. The proposed solution, which consists of a Histogram of Oriented Gradients (HOG) face detector and a set of neural networks, has comparable average accuracy and average true positive and true negative ratio to state-of-the-art deep neural network (DNN) architectures. However, contrary to DNNs, it is possible to easily implement the proposed method in a microcomputer with very limited RAM memory and without the use of additional coprocessors. The proposed method was trained and evaluated on a large 200,000 image face data set and compared with results obtained by other researchers. Further evaluation proves that it is possible to perform facial image attribute classification using the proposed algorithm on incoming video data captured by an RGB camera sensor of the microcomputer. The obtained results can be easily reproduced, as both the data set and source code can be downloaded. Developing and evaluating the proposed facial image annotation algorithm and its implementation, which is easily portable between various hardware and operating systems (virtually the same code works both on high-end PCs and microcomputers using the Windows and Linux platforms) and which is dedicated for low power consumption devices without coprocessors, is the main and novel contribution of this research.

show abstract

“…Object ontology plays an important role in category classification, attribute classification, and conducting training flow and retrieval flow to save computational costs in the training stage and retrieval stage on large-scale datasets, respectively. Thus, based on our experience in researching objects related to attributes such as face [1], cloth [2], person (reidentification), crowd (monitoring) [3, 4], and fast filters in large-scale object retrieval [5], we would like to introduce an object ontology as a hierarchical semantic tree with three levels: region, category, and attribute levels. The attribute level consisted of visual concepts and specific concepts.…”

Section: Introductionmentioning

confidence: 99%

“…Most of the works only present the set of attributes in the form of item lists or item groups [1, 3, 5–10]. A few works used the terminology “ontology” [11], but to the best of our knowledge, there are not works that present the object ontology in full meaning of regions, categories, and attributes.…”

Section: Introductionmentioning

confidence: 99%

Large-Scale Coarse-to-Fine Object Retrieval Ontology and Deep Local Multitask Learning

Kl.²,

Nguyen

2019

Computational Intelligence and Neuroscience

Self Cite

View full text Add to dashboard Cite

Object retrieval plays an increasingly important role in video surveillance, digital marketing, e-commerce, etc. It is facing challenges such as large-scale datasets, imbalanced data, viewpoint, cluster background, and fine-grained details (attributes). This paper has proposed a model to integrate object ontology, a local multitask deep neural network (local MDNN), and an imbalanced data solver to take advantages and overcome the shortcomings of deep learning network models to improve the performance of the large-scale object retrieval system from the coarse-grained level (categories) to the fine-grained level (attributes). Our proposed coarse-to-fine object retrieval (CFOR) system can be robust and resistant to the challenges listed above. To the best of our knowledge, the new main point of our CFOR system is the power of mutual support of object ontology, a local MDNN, and an imbalanced data solver in a unified system. Object ontology supports the exploitation of the inner-group correlations to improve the system performance in category classification, attribute classification, and conducting training flow and retrieval flow to save computational costs in the training stage and retrieval stage on large-scale datasets, respectively. A local MDNN supports linking object ontology to the raw data, and an imbalanced data solver based on Matthews’ correlation coefficient (MCC) addresses that the imbalance of data has contributed effectively to increasing the quality of object ontology realization without adjusting network architecture and data augmentation. In order to evaluate the performance of the CFOR system, we experimented on the DeepFashion dataset. This paper has shown that our local MDNN framework based on the pretrained NASNet architecture has achieved better performance (14.2% higher in recall rate) compared to single-task learning (STL) in the attribute learning task; it has also shown that our model with an imbalanced data solver has achieved better performance (5.14% higher in recall rate for fewer data attributes) compared to models that do not take this into account. Moreover, MAP@30 hovers 0.815 in retrieval on an average of 35 imbalanced fashion attributes.

show abstract

Large-Scale Face Image Retrieval System at Attribute Level Based on Facial Attribute Ontology and Deep Neuron Network

Cited by 12 publications

References 26 publications

An Overview of Facial Attribute Learning

An Overview of Facial Attribute Learning

A Method for Human Facial Image Annotation on Low Power Consumption Autonomous Devices

Large-Scale Coarse-to-Fine Object Retrieval Ontology and Deep Local Multitask Learning

Contact Info

Product

Resources

About