@FJiaojiao Profile picture

Jiaojiao Fan

@FJiaojiao

PhD at Georgia Tech

Joined November 2018
Similar User
Chieh-Hsin (Jesse) Lai photo

@JCJesseLai

Mufan Li photo

@mufan_li

Jia-Jie Zhu (hiring PhD students) photo

@__jzhu__

Antonin Schrab photo

@AntoninSchrab

Mathieu Dagréou photo

@Mat_Dag

Guan-Horng Liu photo

@guanhorng_liu

Daniel Beaglehole photo

@dbeagleholeCS

Dimitri Meunier photo

@DimitriMeunier1

Yongxin Chen photo

@YongxinChen1

Utkarsh Mishra photo

@utkarshm0410

thongnguyen050999 photo

@thongnguyen5999

Thao Nguyen (Shibe) photo

@thaoshibe

Tianyi Chen photo

@Tianyi2020

Khai Nguyen photo

@khainb_ml

Raaz Dwivedi photo

@raazdwivedi

Pinned

Edify image paper is out! "Edify Image: High-Quality Image Generation with Pixel Space Laplacian Diffusion Models" research.nvidia.com/labs/dir/edify…

Tweet Image 1

Jiaojiao Fan Reposted

📗 Project page: research.nvidia.com/labs/dir/edify… 📝 Research paper: arxiv.org/abs/2411.07126 🖼️ Edify Image from #NVIDIAResearch. 🧵 2/2

Tweet Image 1

Jiaojiao Fan Reposted

👀 Pixel perfect 💎✨ 🖼️ Edify Image from #NVIDIAResearch is a family of diffusion models that supports a wide range of applications, including text-to-image synthesis, 4K upsampling, ControlNets, 360° HDR panorama generation, and finetuning for image customization. 🧵 1/2


Jiaojiao Fan Reposted

Nvidia presents Edify Image High-Quality Image Generation with Pixel Space Laplacian Diffusion Models

Tweet Image 1

Jiaojiao Fan Reposted

A simple modification of the attention layers can enhance consistent generations of images and video generative models. Great work done by @FJiaojiao @Haotianxue_GT @qsh_zh

🌟 RefDrop is a plug-and-play,training free approach that can enhance consistent generation for T2I(Row 1)and T2V(Row3) models. It can also do some cool applications like image blending(Row2)!

Tweet Image 1


Jiaojiao Fan Reposted

🌟 RefDrop is a plug-and-play,training free approach that can enhance consistent generation for T2I(Row 1)and T2V(Row3) models. It can also do some cool applications like image blending(Row2)!

Tweet Image 1

🚀 Introducing "RefDrop: Controllable Consistency in Image or Video Generation via Reference Feature Guidance." We present a novel method to control consistency in image/video generation, enabling several interesting applications. Project page: sbyebss.github.io/refdrop/



🚀 Introducing "RefDrop: Controllable Consistency in Image or Video Generation via Reference Feature Guidance." We present a novel method to control consistency in image/video generation, enabling several interesting applications. Project page: sbyebss.github.io/refdrop/


Loading...

Something went wrong.


Something went wrong.