The Real to #Ghibli Image Dataset is a high-quality collection of 5,000 images designed for AI-driven style transfer and artistic transformations. This dataset is ideal for training GANs, CycleGAN, diffusion models, and other deep learning applications in image-to-image translation.It consists of two separate subsets:trainA (2,500 Real-World Images) β A diverse collection of human faces, landscapes, rivers, mountains, forests, buildings, vehicles, and more.
trainB_ghibli (2,500 Ghibli-Style Images) β Stylized images inspired by Studio Ghibli movies, including animated characters, landscapes, and artistic compositions.
Unlike paired datasets, this collection contains independent images in each subset, making it suitable for unsupervised learning approaches.
https://t.me/datasets1
Please open Telegram to view this post
VIEW IN TELEGRAM
β€6π2π₯1
Please open Telegram to view this post
VIEW IN TELEGRAM
π₯7π2
π1π₯1
Forwarded from Machine Learning with Python
This channels is for Programmers, Coders, Software Engineers.
0οΈβ£ Python
1οΈβ£ Data Science
2οΈβ£ Machine Learning
3οΈβ£ Data Visualization
4οΈβ£ Artificial Intelligence
5οΈβ£ Data Analysis
6οΈβ£ Statistics
7οΈβ£ Deep Learning
8οΈβ£ programming Languages
β
https://t.me/addlist/8_rRW2scgfRhOTc0
β
https://t.me/Codeprogrammer
Please open Telegram to view this post
VIEW IN TELEGRAM
β€1
The Ghibli Art Image Dataset is a prototype designed for generating Ghibli-style images using machine learning. It includes three subsetsβtraining, testing, and validationβeach containing directories with two PNG images: one original (o.png) and one generated in Ghibli style (g.png). This dataset supports tasks like image classification and Ghibli-style image generation. As a small-scale version of a larger dataset, it provides essential resources like model code and a pre-trained Generator.pth model. The images were collected from platforms such as Meta, Google, and Instagram for research and experimentation purposes.
https://t.me/datasets1
Please open Telegram to view this post
VIEW IN TELEGRAM
π4β€2
The dataset consists of authentic images sampled from the Shutterstock platform across various categories, including a balanced selection where one-third of the images feature humans. These authentic images are paired with their equivalents generated using state-of-the-art generative models. This structured pairing enables a direct comparison between real and AI-generated content, providing a robust foundation for developing and evaluating image authenticity detection systems.
https://t.me/datasets1
Please open Telegram to view this post
VIEW IN TELEGRAM
π₯3π2β€1
archive.zip.005
1.9 GB
https://t.me/datasets1
Please open Telegram to view this post
VIEW IN TELEGRAM
β€2π1
πWLASL (World Level American Sign Language) Video
β 12k processed videos of Word-Level American Sign Language glossary performance.
π
https://t.me/datasets1π―
β 12k processed videos of Word-Level American Sign Language glossary performance.
π
WLASL is the largest video dataset for Word-Level American Sign Language (ASL) recognition, which features 2,000 common different words in ASL. We hope WLASL will facilitate the research in sign language understanding and eventually benefit the communication between deaf and hearing communities.
https://t.me/datasets1π―
π5π₯2
β€2π2
π
The task of image-based virtual try-on aims to transfer a target clothing item onto the corresponding region of a person, which is commonly tackled by fitting the item to the desired body part and fusing the warped item with the person. While an increasing number of studies have been conducted, the resolution of synthesized images is still limited to low (e.g., 256x192), which acts as the critical limitation against satisfying online consumers. We argue that the limitation stems from several challenges: as the resolution increases, the artifacts in the misaligned areas between the warped clothes and the desired clothing regions become noticeable in the final results; the architectures used in existing methods have low performance in generating high-quality body parts and maintaining the texture sharpness of the clothes. To address the challenges, we propose a novel virtual try-on method called VITON-HD that successfully synthesizes 1024x768 virtual try-on images. Specifically, we first prepare the segmentation map to guide our virtual try-on synthesis, and then roughly fit the target clothing item to a given person's body. Next, we propose ALIgnment-Aware Segment (ALIAS) normalization and ALIAS generator to handle the misaligned areas and preserve the details of 1024x768 inputs. Through rigorous comparison with existing methods, we demonstrate that VITON-HD highly surpasses the baselines in terms of synthesized image quality both qualitatively and quantitatively.https://t.me/datasets1π―
Telegram
Kaggle Data Hub
Your go-to hub for Kaggle datasets β explore, analyze, and leverage data for Machine Learning and Data Science projects.
Admin: @HusseinSheikho || @Hussein_Sheikho
Admin: @HusseinSheikho || @Hussein_Sheikho
π₯2β€1
π₯4