Hypercomplex Image- to- Image Translation

Grassucci, Eleonora; Luigi, Sigillo,; Uncini, Aurelio; Comminiello, Danilo

doi:10.1109/ijcnn55064.2022.9892119

Cited by 6 publications

(6 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…see for example [127], [129], [97], [82], [132], [20], [122], [74], [32], [116], [55], [24]. Another method, proposed by Gaudet and Maida [19], is to use a residual block:…”

Section: A Visionmentioning

confidence: 99%

“…The model is presented in Figure 7c) and 7d). Thereafter, Grassucci et al [132] proposed the quaternion-valued version of the StarGANv2 model [146]. It is composed of the generator, mapping, encoding, and discriminator networks; this model was evaluated on an image to image translation task using the CelebA-HQ dataset [131].…”

Section: A Visionmentioning

confidence: 99%

“…It is composed of the generator, mapping, encoding, and discriminator networks; this model was evaluated on an image to image translation task using the CelebA-HQ dataset [131]. Because of the complexity of the model, it is recommended to consult them directly from [132].…”

Section: A Visionmentioning

confidence: 99%

“…Models Website Gaudet and Maida [19] Residual https://github.com/gaudetcj/DeepQuaternionNetworks/ Grassuccii et al [97], [70], [132], [82] VAE, GAN, Hypercomplex https://github.com/eleGAN23 Parcollet et al [22], [32] QRNNs, QLSTMs, QCNNs (Con-vNets), QCAEs.…”

Section: Authorsmentioning

confidence: 99%

See 3 more Smart Citations

Quaternion Quantum Neural Network for Classification

Altamirano-Escobedo

Bayro–Corrochano

2023

Adv. Appl. Clifford Algebras

View full text Add to dashboard Cite

Since their first applications, Convolutional Neural Networks (CNNs) have solved problems that have advanced the state-of-the-art in several domains. CNNs represent information using real numbers. Despite encouraging results, theoretical analysis shows that representations such as hyper-complex numbers can achieve richer representational capacities than real numbers, and that Hamilton products can capture intrinsic interchannel relationships. Moreover, in the last few years, experimental research has shown that Quaternion-Valued CNNs (QCNNs) can achieve similar performance with fewer parameters than their real-valued counterparts. This paper condenses research in the development of QCNNs from its very beginnings. We propose a conceptual organization of current trends and analyze the main building blocks used in the design of QCNN models. Based on this conceptual organization, we propose future directions of research.

show abstract

“…see for example [127], [129], [97], [82], [132], [20], [122], [74], [32], [116], [55], [24]. Another method, proposed by Gaudet and Maida [19], is to use a residual block:…”

Section: A Visionmentioning

confidence: 99%

Section: A Visionmentioning

confidence: 99%

Section: A Visionmentioning

confidence: 99%

Section: Authorsmentioning

confidence: 99%

See 2 more Smart Citations

Quaternion Quantum Neural Network for Classification

Altamirano-Escobedo

Bayro–Corrochano

2023

Adv. Appl. Clifford Algebras

View full text Add to dashboard Cite

show abstract

“…Furthermore, since the proposed wavelet-based preprocessing is performed in the quaternion domain, we can easily seize quaternion-valued neural networks (QNNs) for the learning stage to further exploit the capabilities of hypercomplex algebra. Indeed, QNNs have been proven to achieve interesting results in processing natural images [27]- [31]. Due to the four-dimensional nature of quaternions, QNNs properly handle inputs with 4 dimensions/channels, thus restricting the use of QNNs to a few sets of data.…”

Section: Introductionmentioning

confidence: 99%

StawGAN: Structural-Aware Generative Adversarial Networks for Infrared Image Translation

Luigi

Grassucci

Comminiello

2023

2023 IEEE International Symposium on Circuits and Systems (ISCAS)

Self Cite

View full text Add to dashboard Cite

Neural network generalizability is becoming a broad research field due to the increasing availability of datasets from different sources and for various tasks. This issue is even wider when processing medical data, where a lack of methodological standards causes large variations being provided by different imaging centers or acquired with various devices and cofactors. To overcome these limitations, we introduce a novel, generalizable, data-and task-agnostic framework able to extract salient features from medical images. The proposed quaternion wavelet network (QUAVE) can be easily integrated with any preexisting medical image analysis or synthesis task, and it can be involved with real, quaternion, or hypercomplex-valued models, generalizing their adoption to single-channel data. QUAVE first extracts different sub-bands through the quaternion wavelet transform, resulting in both low-frequency/approximation bands and high-frequency/fine-grained features. Then, it weighs the most representative set of sub-bands to be involved as input to any other neural model for image processing, replacing standard data samples. We conduct an extensive experimental evaluation comprising different datasets, diverse image analysis, and synthesis tasks including reconstruction, segmentation, and modality translation. We also evaluate QUAVE in combination with both real and quaternion-valued models. Results demonstrate the effectiveness and the generalizability of the proposed framework that improves network performance while being flexible to be adopted in manifold scenarios. The full code is available at: https://github.com/ispamm/QWT.

show abstract

Applications and Limits of Image-to-Image Translation Models

Coscia

Genovese

Scotti

et al. 2023

2023 24th International Conference on Digital Signal Processing (DSP)

View full text Add to dashboard Cite

Image-to-image (121) translation models are widely employed in several fields, e.g., computer vision, security or medicine. Their goal is to map images from a source domain to a target domain while preserving content information.Despite their success, these models suffer from multiple weaknesses. For example, many practical scenarios d o not consent to collect a sufficient amount of images, leading to imbalanced domains. Furthermore, mode collapse and training instability require a careful design and further discourage their deployment o n edge devices. Finally, 121 models need an intensive computation to learn conditional probability distributions and are difficult to adapt to different contexts. These drawbacks mainly limit their large scale applicability.In this work, we want to shed light on the main solutions adopted to overcome the above issues and their impact on the performance.We also investigate several approaches to deploy these models on low-powered devices and weight sharing techniques to reduce the number of parameters and resources used. Index Terms-Image-to-image translation, GAN, cyclic loss I . INTRODUCTION This work was supported i n part b y the EC under project EdgeAI (101097300), b y the Italian MUR under PON project GLEAN, and b y project SERICS (PE00000014) under the MUR NRRP funded b y the EU -NextGenerationEU. Views and opinions expressed are however those o f the authors only and do not necessarily reflect those o f the European Union or the Italian MUR. Neither the European Union nor Italian MUR can be held responsible for them.

show abstract

Hypercomplex Image- to- Image Translation

Cited by 6 publications

References 24 publications

Quaternion Quantum Neural Network for Classification

Quaternion Quantum Neural Network for Classification

StawGAN: Structural-Aware Generative Adversarial Networks for Infrared Image Translation

Applications and Limits of Image-to-Image Translation Models

Contact Info

Product

Resources

About