NEWS: At @CES
, L’Oreal CEO Nicolas Hieronimus just debuted a new “Beauty Genius” AI chatbot to give personalized beauty recommendations to people based on facial recognition and other information. Hieronimus just did a live demo on stage showing how it works. #ces #CES2024
MULTIMODAL AI
-
L’Oreal launches Beauty Genius AI chatbot with facial recognition
By
–
-
DiffMorpher: Experiment with Image Morphing AI Tool
By
–
An experiment with DiffMorpherhttps://t.co/bAkFlPFQOD pic.twitter.com/lQXpSmI68P
— fofr (@fofrAI) 9 janvier 2024An experiment with DiffMorpher https://
replicate.com/p/jyrylydbp3qr
g2gdyfx7yf7hia
… -
New Video Manipulation Tools: Reverse and Bounce Effects
By
–
A couple of small updates to my toolkit model, 2 new tasks:
— fofr (@fofrAI) 9 janvier 2024
– reverse video or gif
– bounce video or gif (ie concatenate the original with the reversed)https://t.co/ozBN8X2SRk
I added this to make it easier to play with outputs from DiffMorpher:https://t.co/Y0sW6dAs0d pic.twitter.com/lCQ5miEI7AA couple of small updates to my toolkit model, 2 new tasks: – reverse video or gif
– bounce video or gif (ie concatenate the original with the reversed) https://
replicate.com/p/ekqqlezbadcw
qahtguejo7zu7m
… I added this to make it easier to play with outputs from DiffMorpher: https://
replicate.com/cjwbw/diffmorp
her
… -
PhD Research in Computer Vision and Embodied AI Navigation
By
–
Links to these brilliant PhD candidates' work: Sonia Raychauduri https://
openaccess.thecvf.com/content/WACV20
24/html/Raychaudhuri_MOPA_Modular_Object_Navigation_With_PointGoal_Agents_WACV_2024_paper.html
… Bita Azari https://
openaccess.thecvf.com/content/WACV20
24/html/Azari_EmoStyle_One-Shot_Facial_Expression_Editing_Using_Continuous_Emotion_Parameters_WACV_2024_paper.html
… Sha Hu https://
openaccess.thecvf.com/content/WACV20
24/html/Hu_Embodied_Human_Activity_Recognition_WACV_2024_paper.html
… -
Denoising Vision Transformers: Advanced Deep Learning Research
By
–
Denoising Vision Transformers Yang et al.: https://
arxiv.org/abs/2401.02957 #ArtificialIntelligence #DeepLearning #MachineLearning -
AI Model Demonstrates Human-Like Visual Attention Mechanisms
By
–
The way it attends to relevant parts of the visual input seems almost human, though its attending in parallel as well as sequentially.
-
Multi-Model Image Generation with Aspect Ratio Selection
By
–
4️⃣ Multi-modal Generation with aspect ratio selection:
— Pietro Schirano (@skirano) 8 janvier 2024
Select up to 5 different models at once to generate images simultaneously. It's perfect for comparing styles or seeking inspiration when you need it most. First of its kind on @everartai pic.twitter.com/5lc4cbkZHPMulti-modal Generation with aspect ratio selection: Select up to 5 different models at once to generate images simultaneously. It's perfect for comparing styles or seeking inspiration when you need it most. First of its kind on @everartai
-
Pose Transfer: Transform Images Into Your Custom Style
By
–
1️⃣ Pose Transfer: Ever seen an image and wished it could come to life in your style? @everartai can do it.
— Pietro Schirano (@skirano) 8 janvier 2024
Just select an image and watch as it transforms into any style you've created. Ideal for consistent character poses, especially with models trained on people. pic.twitter.com/lFsnp6215VPose Transfer: Ever seen an image and wished it could come to life in your style? @everartai can do it. Just select an image and watch as it transforms into any style you've created. Ideal for consistent character poses, especially with models trained on people.
-
Magic Prompting: Transform Ideas into Stunning AI Art
By
–
3️⃣ Magic Prompting: From Thought to Art 🧙♂️
— Pietro Schirano (@skirano) 8 janvier 2024
Struggling with ideas? Write a simple thought and @everartai suggests 5 incredible prompts to spark your creativity and create stunning images with. pic.twitter.com/KqjtPwK3oFMagic Prompting: From Thought to Art Struggling with ideas? Write a simple thought and @everartai suggests 5 incredible prompts to spark your creativity and create stunning images with.
-
DocGraphLM: Document Graph Language Model for Information Extraction
By
–
DocGraphLM: Documental Graph Language Model for Information Extraction Wang et al.: https://
arxiv.org/abs/2401.02823 #ArtificialIntelligence #DeepLearning #MachineLearning