🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

AI that uses sketches to detect objects within an image could boost tumour detection, and search for rare bird species 🐦🐦🐦 surrey.ac.uk/news/ai-uses-s… Yi-Zhe Song SketchX Surrey Institute for People-Centred AI

The Academic Surgical Unit met with University of Surrey on Thursday. Thanks to Prof Ferrante Neri, Prof Yi-Zhe Song and team for a great meeting. Professor Deiary Kader (Medical Doctor) Irrum Afzal Sarkhell Radha Vipin Asopa #artificalintelligence #AIinOrthopaedics

🎉 Amazing news! Our paper "What Can Human Sketches Do for Object Detection?" is featured in RSIP Vision's BEST OF CVPR! 📚 Check it out on page 2: rsipvision.com/ComputerVision… 🔍 More about this project: pinakinathc.me/sketch-detect #CVPR #ResearchHighlights SketchX Surrey Institute for People-Centred AI

Very proud of University of Surrey community’s continuing success in terms of outstanding teaching and student experience, as validated by the results in NSS2023. Well done and thank-you to everyone for your hard work and commitment. surrey.ac.uk/news/universit…

To seize the transformative benefits of AI we need more people studying & researching it at our top unis, like University of Surrey where I'm seeing cutting edge AI work today

Amazing work by Zhiyu Qu, Yi-Zhe Song & others from SketchX on Multi-view Wire Art. They optimise a set of 3D wires such that projection on 3 orthogonal surfaces match given prompts. #generativeart #generativeAI #AIart #sketchart 1/3 🧵

🚨DemoFusion: Cool new work by Ruoyi Du, Dongliang Chang, Yi-Zhe Song and others that leverage existing Diffusion Models to produce images of increasingly high resolution, without training. Work done in colab with PRIS Lab BUPT, Tsinghua University, The University of Edinburgh and SketchX. 1/4 🧵

The video 📽️ shows more qualitative comparisons between DemoFusion and other competing methods. 2/4 🧵 #generativeAI #ImageGeneration #diffusionmodels

Excited to have been part of DemoFusion, bringing UHD generation to SDXL on your desktop with no training! With Ruoyi Du Yi-Zhe Song Dongliang Chang Project: ruoyidu.github.io/demofusion/dem…, paper: arxiv.org/abs/2311.16973 #GenerativeAI

A free-to-use #AI tool can make images as detailed as the Santa on the right - using no more computing power than a mid-range gaming PC. Researchers at our Surrey Institute for People-Centred AI institute hope DemoFusion will open up access to hi res AI image generation. ow.ly/Av7750Qf1aO

The University of Surrey has unveiled a new AI image generator, which builds on open-source models and then enhances resolution thenextweb.com/news/new-ai-to…

💰DemoFusion: High-resolution generation using only SDXL and a RTX 3090 GPU! ... is now available in 🧨diffusers as a community pipeline! Check it out: github.com/huggingface/di… Project Page: ruoyidu.github.io/demofusion/dem… #generativeAI #ImageGeneration #diffusionmodels

Gradio demo is now available at our Repo! Better interaction! Local demo: github.com/PRIS-CV/DemoFu….

Here's the demo "Enhance This"! It's a surreal image magnifier that creates a high-res version by imagining new details, using the SDXL base model. Thanks to Ruoyi Du's DemoFusion research. It takes a ~minute to generate a 2024x2024 image. huggingface.co/spaces/radames…

run with docker: huggingface.co/spaces/radames…

I had the chance to play with a new model built on top of SDXL earlier this week. DemoFusion takes multiple pass throughs, transferring noise data to each pass in order to generate high resolution images up to 16x the default. Read my report here for Tom’s Guide:

🔬 Our study introduces "Upscale-A-Video," a text-guided latent diffusion framework for video upscaling. It ensures temporal coherence locally & globally, balancing fidelity and quality. 🚀 Project page: shangchenzhou.com/projects/upsca… 💻 GitHub: github.com/sczhou/Upscale… 🎥 Video:

Thanks for picking this up No Film School! Live demo coming up!

#BMVC2024 Calls for Paper and timeline now are available! • Abstract submission: April 26, 2024 • Paper submission: May 3, 2024 • Acceptance notification: June 19, 2024 • Conference: November 25-28, 2024 More details can be checked on bmvc2024.org. Save the dates!

Happy Friday!!