Generic image application using GANs (Generative Adversarial Networks): A Review

Porkodi, S; Sarada, V.; Maik, Vivek; Gurushankar, K.

doi:10.1007/s12530-022-09464-y

Cited by 16 publications

(5 citation statements)

References 58 publications

(57 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This study is going to deal with GANs, and as stated in the introduction, Goodfellow et al [68]. Created generative adversarial networks (GANs), a kind of deep learning model, in 2014.…”

Section: B 2d Model Generationmentioning

confidence: 99%

PIXGAN-Drone: 3D Avatar of Human Body Reconstruction From Multi-View 2D Images

Salim Rasheed,

Jabberi,

Hamdani

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Research on training Generative Adversarial Networks (GANs) to create 3D human body avatars from 2D datasets is underway. Research done in this field shows promise and has paved the way for significant advancements in a variety of applications, including virtual reality, sports analysis, cinematography, surveillance, and cinematography. By avoiding obstacles and producing high-resolution, rich-information multi-view (RGB) images, drone active tracking combined with aerial photography sensors can eliminate occlusions and enable 3D avatar body reconstruction. Due to several issues, such as restricted perspective coverage, obvious occlusions, and texture disappearance, 3D avatar reconstruction techniques encounter training failures that cause distortions and feature loss in 3D reconstructed models. The new endto-end trainable deep neural network methodology PIXGAN-Drone is presented in this paper for the photorealistic 3D avatar of the human body reconstruction from multi-view images. It is based on the integration of active tracking drones equipped with aerial photography sensors (stable automatic circular motion system) into the Pix2Pix GANs training framework to generate high-resolution 2D models. Conditional image-toimage translation and dynamic aerial perspectives can be used to develop realistic and accurate 3D models. This research conducted experiments on multiple datasets to demonstrate the improved performance of our method over state-of-the-art methods for various metrics (Chamfer, P2S, and CED). The results showed that our 3D reconstructed human avatars were 0.0293, 0.0271, and 0.0232 on RenderPeople, 0.0133, 0.0136, 0.0050 on People Snapshot (indoor), 0.0154, 0.0101, 0.0063 on People Snapshot (outdoor), and 0.0316, 0.0275, 0.0216 on Custom data-drone (collected dataset).INDEX TERMS 3D human avatar, 3D reconstruction, PIX2PIXGAN, body model rendering, drone active tracking.

show abstract

“…This study is going to deal with GANs, and as stated in the introduction, Goodfellow et al [68]. Created generative adversarial networks (GANs), a kind of deep learning model, in 2014.…”

Section: B 2d Model Generationmentioning

confidence: 99%

PIXGAN-Drone: 3D Avatar of Human Body Reconstruction From Multi-View 2D Images

Salim Rasheed,

Jabberi,

Hamdani

et al. 2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Nevertheless, GANs have their drawbacks, including stability issues and the complexity of training. While they exhibit impressive performance in amodal appearance reconstruction, their usage in amodal segmentation and order recovery tasks is comparatively limited [48]. Despite these challenges, leveraging GANs for occlusion handling remains pivotal in computer vision applications.…”

Section: Generative Adversarial Network Approachmentioning

confidence: 99%

Enhancing Object Detection in Smart Video Surveillance: A Survey of Occlusion-Handling Approaches

Ouardirhi,

Mahmoudi,

Zbakh

2024

Electronics

View full text Add to dashboard Cite

Smart video surveillance systems (SVSs) have garnered significant attention for their autonomous monitoring capabilities, encompassing automated detection, tracking, analysis, and decision making within complex environments, with minimal human intervention. In this context, object detection is a fundamental task in SVS. However, many current approaches often overlook occlusion by nearby objects, posing challenges to real-world SVS applications. To address this crucial issue, this paper presents a comprehensive comparative analysis of occlusion-handling techniques tailored for object detection. The review outlines the pretext tasks common to both domains and explores various architectural solutions to combat occlusion. Unlike prior studies that primarily focus on a single dataset, our analysis spans multiple benchmark datasets, providing a thorough assessment of various object detection methods. By extending the evaluation to datasets beyond the KITTI benchmark, this study offers a more holistic understanding of each approach’s strengths and limitations. Additionally, we delve into persistent challenges in existing occlusion-handling approaches and emphasize the need for innovative strategies and future research directions to drive substantial progress in this field.

show abstract

“…Additionally, it addresses concerns about the potential misuse of GANs for creating fraudulent images and videos and highlights initiatives aimed at combating such misuse. [5] The paper presents the use of supervised machine learning (ML) algorithms to classify brain MRI reports and identify patients with acute ischemic stroke (AIS). The study involved analyzing 3,204 brain MRI documents, of which 432 (14.3%) were labeled as AIS.…”

Section: Literature Surveymentioning

confidence: 99%

Brain Stroke Detection Through Gan-Ml Integration

2024

IRJMETS

View full text Add to dashboard Cite

Brain strokes are a significant global health concern and the second leading cause of death worldwide. In India, it's reported that three people suffer from a brain stroke every minute. There are two main types of brain strokes: ischemic and hemorrhagic strokes. Ischemic strokes, which occur when blood clots block an artery in the brain, are the most common, accounting for around 80% of cases. Hemorrhagic strokes, caused by the rupture of a blood vessel in the brain, make up the remaining 20%. The GAN-ML integration approach focuses on the generation and analysis of medical images that help in the detection and classification of brain strokes. Generative adversarial networks are a subset of artificial intelligence that excels at generating and processing images and data. GAN can be used to synthesize diverse sets of brain images, including normal and strokes. This synthetic data is used in training ML models, which can learn to identify patterns and anomalies in brain strokes. The model is designed based on a deep neural network, including dataset collection, preprocessing, and feature extraction, with the desired model and classification. In this process, using GAN to create diverse data samples augments the dataset for ML models. By learning from both real and synthetic data, ML models become more accurate in distinguishing healthy brain scans from those showing signs of a stroke.

show abstract

Generic image application using GANs (Generative Adversarial Networks): A Review

Cited by 16 publications

References 58 publications

PIXGAN-Drone: 3D Avatar of Human Body Reconstruction From Multi-View 2D Images

PIXGAN-Drone: 3D Avatar of Human Body Reconstruction From Multi-View 2D Images

Enhancing Object Detection in Smart Video Surveillance: A Survey of Occlusion-Handling Approaches

Brain Stroke Detection Through Gan-Ml Integration

Contact Info

Product

Resources

About