Author: Ekrem Çetinkaya

126 POSTS0 COMMENTS
Ekrem Çetinkaya received his B.Sc. in 2018, and M.Sc. in 2019 from Ozyegin University, Istanbul, Türkiye. He wrote his M.Sc. thesis about image denoising using deep convolutional networks. He received his Ph.D. degree in 2023 from the University of Klagenfurt, Austria, with his dissertation titled "Video Coding Enhancements for HTTP Adaptive Streaming Using Machine Learning." His research interests include deep learning, computer vision, video encoding, and multimedia networking.

Point-Cloud Completion with Pretrained Text-to-image Diffusion Models

Have you ever heard the term point-cloud? It is a fundamental representation of 3D data, consisting of points in a three-dimensional coordinate system that...

The Sculpture of Dreams: DreamTime is An AI Model That Improves the Optimization Strategy for Text-to-3D Content Generation

Generative AI models are now a part of our daily lives. They have advanced rapidly in recent years, and the results went from a...

Meet DiffusionDet: An Artificial Intelligence (AI) Model That Uses Diffusion for Object Detection

Object detection is a powerful technique for identifying objects in images and videos. Thanks to deep learning and computer vision advances, it has come...

What Did You Feed On? This AI Model Can Extract Training Data From Diffusion Models

Diffusion models became a key part of the AI domain in 2022. We have seen photorealistic images generated by them, and they kept getting...

Dive Thinking Like an Annotator: Generation of Dataset Labeling Instructions

We are all amazed by the advancement we have seen in AI models recently. We’ve seen how generative models revolutionized themselves by going from...

Segment Anything, but Faster! This AI Approach Speeds Up the SAM Model

Finding objects in images has been a long-going task in computer vision. Object detection algorithms try to locate the objects by drawing a box...

No, no, Let’s Not Put it There! This AI Method Can Do Continuous Layout Editing with Diffusion Models

At this point, everyone is familiar with text-to-image models. They made their way in with the release of stable diffusion last year, and since...

Transcending Into Consistency: This AI Model Teaches Diffusion Models 3D Awareness for Robust Text-to-3D Generation

Text-to-X models have grown rapidly recently, with most of the advancement being in text-to-image models. These models can generate photo-realistic images using the given...

Metrics Can Deceive, but Eyes Can Not: This AI Method Proposes a Perceptual Quality Metric for Video Frame Interpolation

The advancement in display technology has made our viewing experience more intense and pleasant. Watching something in 4K 60FPS is extremely satisfying than 1080P...

Think Like this and Answer Me: This AI Approach Uses Active Prompting to Guide Large Language Models

We all got familiar with large language models (LLMs) in recent months with the introduction of ChatGPT, and it quickly became an essential tool...

The Magic Brush That Works in 3D: Blended-NeRF is an AI Model That Does Zero-Shot Object Generation in Neural Radiance Fields

The recent couple of years were full of eureka moments for various disciplines. We have witnessed revolutionary methods emerging that resulted in colossal advancements....

Do Modern Problems Require Modern Solutions? This Study Examines the Effectiveness of MLPs (Multi-Layer Perceptions) in Modern Settings

Deep Learning is everywhere. It has revolutionized lots of fields, and the trend seems to continue in the near future. Deep learning is a...

Check Out Our Super Cool AI Research Newsletter While It's Still Free

X