Integrated generalized zero-shot learning for fine-grained classification

Shermin, Tasfia; Teng, Shyh Wei; Sohel, Ferdous; Murshed, Manzur; Lu, Guojun

doi:10.1016/j.patcog.2021.108246

Cited by 22 publications

(4 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…At last, we evaluate SAT on object detection and identify images of bicycle and electric bicycle. According to our training protocol, SAT model appears perfect performance, such as mAP (Albuquerque et al,2021) and maxDets (Shermin et al,2021). The Average Precision (AP) reaches 95%, the small area metric reaches 90%, the medium area metric reaches 94.8% and the large area metric reaches 94.5%, as shown in Table 4.…”

Section: Object Detectionmentioning

confidence: 83%

An Enhanced Light Object Detection for Indiscernible Object in the Special Scene

Zhang

Feng²,

Wang³

et al. 2022

JER is an international, peer-reviewed journal that publishes f

View full text Add to dashboard Cite

We present a transfer learning method named Special Application Transfer (SAT) for special object detection in a real life scenario. Our method improves fine-tuning hyper-parameter and adds unrecognized samples to detect special samples when training object detection neural networks for classification. We implement the model of NanoDet on special supervised datasets and fine-tune the hyper-parameter on a target task. More importantly, we combine a few carefully selected samples in training and simple heuristic fine-tuning to achieve good performance on special object detection in real-life scenarios. Our method (SAT) performs well across surprisingly the small dataset the medium dataset and the large dataset. SAT achieves 95% AP (Average Precision) on the small dataset, 94.8% AP on the medium dataset, and 94.5% AP on the large dataset. The performances of AP run-time and training convergence are perfect, compared with the original method and well-established famous methods on the challenging COCO dataset and our dataset. We hope our work could promote and complete the practical application in more real life scenarios.

show abstract

Section: Object Detectionmentioning

confidence: 83%

An Enhanced Light Object Detection for Indiscernible Object in the Special Scene

Zhang

Feng²,

Wang³

et al. 2022

JER is an international, peer-reviewed journal that publishes f

View full text Add to dashboard Cite

show abstract

“…This model is trained to master conditional probability distribution; P(y|x, T). This learning type is suitable for machine translation, where the label is absent in the target language [244][245][246].…”

Section: Zero-shot Learningmentioning

confidence: 99%

A survey on deep learning tools dealing with data scarcity: definitions, challenges, solutions, tips, and applications

et al. 2023

View full text Add to dashboard Cite

Data scarcity is a major challenge when training deep learning (DL) models. DL demands a large amount of data to achieve exceptional performance. Unfortunately, many applications have small or inadequate data to train DL frameworks. Usually, manual labeling is needed to provide labeled data, which typically involves human annotators with a vast background of knowledge. This annotation process is costly, time-consuming, and error-prone. Usually, every DL framework is fed by a significant amount of labeled data to automatically learn representations. Ultimately, a larger amount of data would generate a better DL model and its performance is also application dependent. This issue is the main barrier for many applications dismissing the use of DL. Having sufficient data is the first step toward any successful and trustworthy DL application. This paper presents a holistic survey on state-of-the-art techniques to deal with training DL models to overcome three challenges including small, imbalanced datasets, and lack of generalization. This survey starts by listing the learning techniques. Next, the types of DL architectures are introduced. After that, state-of-the-art solutions to address the issue of lack of training data are listed, such as Transfer Learning (TL), Self-Supervised Learning (SSL), Generative Adversarial Networks (GANs), Model Architecture (MA), Physics-Informed Neural Network (PINN), and Deep Synthetic Minority Oversampling Technique (DeepSMOTE). Then, these solutions were followed by some related tips about data acquisition needed prior to training purposes, as well as recommendations for ensuring the trustworthiness of the training dataset. The survey ends with a list of applications that suffer from data scarcity, several alternatives are proposed in order to generate more data in each application including Electromagnetic Imaging (EMI), Civil Structural Health Monitoring, Medical imaging, Meteorology, Wireless Communications, Fluid Mechanics, Microelectromechanical system, and Cybersecurity. To the best of the authors’ knowledge, this is the first review that offers a comprehensive overview on strategies to tackle data scarcity in DL.

show abstract

“…Various modalities have been used as auxiliary information: word embeddings (Frome et al, 2013;Xian et al, 2016), hierarchical embeddings (Kampffmeyer et al, 2019), attributes (Farhadi et al, 2009;Akata et al, 2015) or Wikipedia articles (Elhoseiny et al, 2017;Zhu et al, 2018;Elhoseiny et al, 2016;Qiao et al, 2016). Most recent work uses generative models conditioned on class descriptions to synthesize training examples for unseen categories (Long et al, 2017;Kodirov et al, 2017;Felix et al, 2018;Xian et al, 2019;Vyas et al, 2020;, attention-enabled feature extractors (Yu et al, 2018;Zhu et al, 2019;Shermin et al, 2022;Chen et al, 2022). The multi-modal and often finegrained nature of the standard and generalised (G)ZSL task renders it related to our problem.…”

Section: Zero/few Shot Learningmentioning

confidence: 99%

The Curious Layperson: Fine-Grained Image Recognition Without Expert Labels

Choudhury,

Laina,

Rupprecht

et al. 2023

Int J Comput Vis

View full text Add to dashboard Cite

Most of us are not experts in specific fields, such as ornithology. Nonetheless, we do have general image and language understanding capabilities that we use to match what we see to expert resources. This allows us to expand our knowledge and perform novel tasks without ad-hoc external supervision. On the contrary, machines have a much harder time consulting expert-curated knowledge bases unless trained specifically with that knowledge in mind. Thus, in this paper we consider a new problem: fine-grained image recognition without expert annotations, which we address by leveraging the vast knowledge available in web encyclopedias. First, we learn a model to describe the visual appearance of objects using non-expert image descriptions. We then train a fine-grained textual similarity model that matches image descriptions with documents on a sentence-level basis. We evaluate the method on two datasets (CUB-200 and Oxford-102 Flowers) and compare with several strong baselines and the state of the art in cross-modal retrieval. Code is available at: https://github.com/subhc/clever.

show abstract

Integrated generalized zero-shot learning for fine-grained classification

Cited by 22 publications

References 24 publications

An Enhanced Light Object Detection for Indiscernible Object in the Special Scene

An Enhanced Light Object Detection for Indiscernible Object in the Special Scene

A survey on deep learning tools dealing with data scarcity: definitions, challenges, solutions, tips, and applications

The Curious Layperson: Fine-Grained Image Recognition Without Expert Labels

Contact Info

Product

Resources

About