Imagine "circle to search" gesture and ppl on streets casting it evrywhere
MULTIMODAL AI
-
生成式AI创造中文字符的潜在空间插值
By
–
Next, I held the latents fixed while the character label is gradually interpolated, allowing us to visualize characters *in between* characters, exploring the conceptual space of Chinese and materialize imaginary characters that never were. pic.twitter.com/vOIMiEoMBj
— Gene Kogan (@genekogan) 10 juin 2024Next, I held the latents fixed while the character label is gradually interpolated, allowing us to visualize characters *in between* characters, exploring the conceptual space of Chinese and materialize imaginary characters that never were.
-
Training Character Image Generation Models: Finding the Right Setup
By
–
My first goal was to simply generate still images of characters. It took a handful of training runs to find the right setup. It had not been obvious to me that it would work at all. This was eons before techniques like LoRA made finetuning on small and arbitrary datasets easy.
-
Kling: Kuaishou’s Chinese Text-to-Video Model
By
–
[#Article] Text-to-Video Models: Kling, Kuaishou's Chinese Model, Enters the Competition https://actuia.com/actualite/modeles-text-to-video-kling-le-modele-chinois-de-kuaishou-entre-en-lice/
… #AI #ArtificialIntelligence -
Alchemist: New Diffusion Model Transforms Image Material Properties
By
–
Ever wondered how AI can be used to edit photos? Alchemist is a new diffusion model that can transform the look and feel of objects in images. Read more to see how it works: https://
indiaai.gov.in/article/new-di
ffusion-model-alchemist-transforms-material-properties-in-images
… @MIT_CSAIL @Jeevprabnivash #AI #ArtificialIntelligence #AIdevelopment -
Grok AI adds Socrates mode and teases image generation features
By
–
Read more https://
testingcatalog.com/icymi-grok-int
rospects-with-socrates-mode-and-teases-image-generation-and-web-search-features/
… -
Generative ConvNets for Video: A Decade of AI Innovation
By
–
Us indeed! A decade earlier, in 2010, we organically introduced generative convnets for video understanding and generation. I remember presenting it at NeurIPS and Jitendra Malik rightfully asking us to scale and evaluate. Yet, I argued against most of the vision community at the
-
Video Generation Models Progress Beyond Current Limitations
By
–
The progress in video generation is astounding. Sora, Veo, Kling, etc are only the GPT2s of video at best. When the GPT4 equivalents arise, machines will be able to do reasoning with video, chain-of-thought with video, multimodal generalisation, inference and control from
-
Practical guide to editing images using ChatGPT
By
–
Steps to follow:
— God of Prompt (@godofprompt) 9 juin 2024
1. Go to https://t.co/4M8XLaa8Ih
2. Write the prompt and say something to add text to the image
3. Edit and download pic.twitter.com/W4F3DzUi6DSteps to follow: 1. Go to http://
ChatGPT.com
2. Write the prompt and say something to add text to the image
3. Edit and download