Here's an original image that I created: And here's the same image Upscaled (Creative): The difference isn't massive. It's not like Magnific where you'll see a lot of whole new details added in (yet). I'm super excited about MJ6 but it doesn't feel to me as big of a leap that
MULTIMODAL AI
-
Google Magic Eraser Enhanced with Generative AI Inpainting
By
–
Magic Eraser now uses gen AI to fill in detail when users remove unwanted objects from photos. Google Research worked on the MaskGIT generative image transformer for inpainting, and improved segmentation to include shadows and objects attached to people. pic.twitter.com/GoVtZTxT0y
— Google AI (@GoogleAI) 21 décembre 2023Magic Eraser now uses gen AI to fill in detail when users remove unwanted objects from photos. Google Research worked on the MaskGIT generative image transformer for inpainting, and improved segmentation to include shadows and objects attached to people.
-
AI-Generated Hyperrealistic Pope Portrait Showcases Advanced Image Synthesis
By
–
This was the prompt: “Pope Francis wearing a Balenciaga puffer jacket , taken using a Canon EOS R camera with a 50mm f/1.8 lens, f/2.2 aperture, shutter speed 1/200s, ISO 100 and natural light, Full Body, Hyper Realistic Photography, Cinematic, Cinema, Hyperdetail, UHD, Color
-
Pro Users Now Upload Images Multimodal AI Web Mobile
By
–
We’re excited to extend multimodal capabilities, and make image uploads available to our Pro users on web and mobile! You can upload an image, be it a historic landmark, a puzzle, or a scene from daily life, and ask for explanations! Works especially well with Copilot toggle on. pic.twitter.com/FlamrxkNpp
— Perplexity (@perplexity_ai) 21 décembre 2023We’re excited to extend multimodal capabilities, and make image uploads available to our Pro users on web and mobile! You can upload an image, be it a historic landmark, a puzzle, or a scene from daily life, and ask for explanations! Works especially well with Copilot toggle on.
-
Midjourney V6 Alpha Testing Launches with Improved Coherence
By
–
We're now alpha-testing our V6 models Midjourney. Just type /settings and click V6 or add –v 6 after your prompt. Image coherence and prompt understanding are greatly improved. You can draw text and dolphins and there's new upscalers too. Happy holidays everyone!
-
Getting Started with Visual Assistants: Templates and Video Overview
By
–
Getting started with visual assistants: templates + video overview Visual assistants will be an important theme in 2024 as multi-modal LLMs gain wider adoption and more capabilities. In case you want to explore multi-modal LLMs over the holidays: we've released 5 new
-
Improving VQA Evaluation Using Large Language Models
By
–
Improving Automatic VQA Evaluation Using Large Language Models Mañas et al.: https://
arxiv.org/abs/2310.02567 #DeepLearning #ChatGPT #LargeLanguageModels -
RealVisXL v3 Launched on Replicate for Photorealistic Image Generation
By
–
I've pushed RealVisXL v3 to Replicate: https://
replicate.com/fofr/realvisxl
-v3
… RealVisXL is based on SDXL and is excellent at photorealism. You can fine-tune it too. -
Composite Multiple Gen-2 Videos into Single Scene
By
–
Composite multiple Gen-2 videos into a single scene.
— Runway (@runwayml) 21 décembre 2023
Learn how with today's Runway Academy: https://t.co/Wj74vDpPKo pic.twitter.com/qAS9EPBI1DComposite multiple Gen-2 videos into a single scene. Learn how with today's Runway Academy: https://
academy.runwayml.com/gen2/gen2-comp
ositing-workflow
… -
Real-time speech-to-speech translation preserving speaker voice
By
–
Real-time , low-latency, speech-to-speech translation that preserves the voice and expression of the speaker. https://t.co/eB9qIxr7rp
— Yann LeCun (@ylecun) 21 décembre 2023Real-time , low-latency, speech-to-speech translation that preserves the voice and expression of the speaker.