MIC: Multi-view Image Classifier using Generative Adversarial Networks for Missing Data Imputation

Aversano, Gianmarco; Jarraya, Mahmoud; Marwani, Maher; Lahouli, Ichraf; Skhiri, Sabri

doi:10.1109/ssd52085.2021.9429478

Cited by 5 publications

(5 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although several methods have been proposed to tackle multi-view image classification [1], [6], only a few works have investigated and conceived approaches to handle missing data [5], [7], [8], [9], [10], [11].…”

Section: Related Workmentioning

confidence: 99%

“…More recently, Generative Adversarial Networks (GANs) [13] have gained popularity in the missing data completion field, due to their ability to generate synthetic samples. Although there are several works [8], [10], [11] handling multiview missing data completion using GANs, those are not thoroughly discussed here given that, as introduced, they are outside of the scope of this work, which focuses on information extracted from real (non-synthetic) images. The main reason for this is the fact that training models to generate good quality synthetic data is troublesome mainly when having a small amount of data, a common case in multiview scenarios.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Facing the Void: Overcoming Missing Data in Multi-View Imagery

et al. 2023

View full text Add to dashboard Cite

In some scenarios, a single input image may not be enough to allow the object classification. In those cases, it is crucial to explore the complementary information extracted from images presenting the same object from multiple perspectives (or views) in order to enhance the general scene understanding and, consequently, increase the performance. However, this task, commonly called multi-view image classification, has a major challenge: missing data. In this paper, we propose a novel technique for multi-view image classification robust to this problem. The proposed method, based on state-of-the-art deep learning-based approaches and metric learning, can be easily adapted and exploited in other applications and domains. A systematic evaluation of the proposed algorithm was conducted using two multi-view aerial-ground datasets with very distinct properties. Results show that the proposed algorithm provides improvements in multiview image classification accuracy when compared to state-of-the-art methods. The code of the proposed approach is available at https://github.com/Gabriellm2003/remote_sensing_missing_data.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Facing the Void: Overcoming Missing Data in Multi-View Imagery

et al. 2023

View full text Add to dashboard Cite

show abstract

“…Although several methods have been proposed to tackle multi-view image classification [4,5], only a few works have investigated and conceived approaches to handle missing data [6,7,8,3,9,10].…”

Section: Related Workmentioning

confidence: 99%

“…In this way, a single GAN network can successfully generate the missing data using the remaining (complete) data set. Another GAN-based work was proposed by Aversano et al [10]. In this work, a multi-branch network jointly encodes data of different modalities/views, generating a common embedding space.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Facing the Void: Overcoming Missing Data in Multi-View Imagery

Machado¹,

Nogueira²,

Pereira³

et al. 2022

Preprint

View full text Add to dashboard Cite

In some scenarios, a single input image may not be enough to allow the object classification. In those cases, it is crucial to explore the complementary information extracted from images presenting the same object from multiple perspectives (or views) in order to enhance the general scene understanding and, consequently, increase the performance. However, this task, commonly called multi-view image classification, has a major challenge: missing data. In this paper, we propose a novel technique for multi-view image classification robust to this problem. The proposed method, based on state-of-the-art deep learning-based approaches and metric learning, can be easily adapted and exploited in other applications and domains. A systematic evaluation of the proposed algorithm was conducted using two multi-view aerial-ground datasets with very distinct properties. Results show that the proposed algorithm provides improvements in multi-view image classification accuracy when compared to state-of-the-art methods. Code available at https://github.com/Gabriellm2003/remote_sensing_missing_data.

show abstract

Incomplete multi-view learning: Review, analysis, and prospects

Tang,

Yi,

et al. 2024

Applied Soft Computing

View full text Add to dashboard Cite

MIC: Multi-view Image Classifier using Generative Adversarial Networks for Missing Data Imputation

Cited by 5 publications

References 7 publications

Facing the Void: Overcoming Missing Data in Multi-View Imagery

Facing the Void: Overcoming Missing Data in Multi-View Imagery

Facing the Void: Overcoming Missing Data in Multi-View Imagery

Incomplete multi-view learning: Review, analysis, and prospects

Contact Info

Product

Resources

About