X / Twitter
🆕✨HivisionIDPhoto provides a practical and intelligent ID photo production framework.
Uses a set of models and workflows for portrait recognition, image cutout & ID photo generation for a variety of photography situations.
- Very lightweight cutout generation🚀 ,
- Generate different standard ID photos, and
- Six-inch layout photos according to different sizes and specifications
- Team is working on implementing Beauty focussed release ++ a VTON based smart attire change🤩
I'm excited to share several repos, datasets and models for SAEs on sentence embeddings!
First: a visual interface to the features learned by an SAE, inspired by @neuronpedia
search and browse features, see top activating examples and find similar features
links below https://t.co/L9c0VXxby3
📔 How to fine-tune an LLM locally in 1 minute
💭 using Roller Coaster Tycoon peep thoughts as an example https://t.co/cXfxjrPYIj
🎉 Introducing OmniRe: A 3DGS framework to reconstruct dynamic urban scenes with (almost) all actors (vehicles, pedestrians, cyclists, etc.) in one stage! Extended to datasets: Waymo, PandaSet, Argoverse2, KITTI, NuScenes, Nuplan.
Paper & code see: https://t.co/C8Ic5pzo9E https://t.co/ModNZHSr5U
Want happier, healthier kids?
Offer them autonomy.
Massive new meta-analysis makes it plain:
In every culture, control leaves children worse off — but autonomy helps them people engage, learn, and grow. https://t.co/XmbVeLlHZ9
🤯🤯 Kotaemon - An open-source clean & customizable RAG UI, built with Gradio, for chatting with your docs. A UI built with both end users and developers in mind.🧡 https://t.co/vphUIIfb89
The hybrid SSM scene is on 🔥 - Zyphyra 1.2B - Apache 2.0 licensed, a strong base model!
> Beats Gemma 2B, Phi 1.5, OpenELM, etc
> Trained on 3T tokens (annealed w/ 100B high-quality tokens)
> Mamba 2 + a shared transformer block
> LoRA projectors to attention blocks (instead of just on MLP block)
> RoPE
> Uses Mistral 7B v 0.1 tokenizer
> Works w/ Transformers (PR)
> Model checkpoints on the Hub 🤗
Congrats, @ZyphraAI team - I looking forward to the instruction-tuned versions and further releases!
LayerPano3D
Layered 3D Panorama for Hyper-Immersive Scene Generation
discuss: https://t.co/ibqSFaB74o
3D immersive scene generation is a challenging yet critical task in computer vision and graphics. A desired virtual 3D scene should 1) exhibit omnidirectional view consistency, and 2) allow for free exploration in complex scene hierarchies. Existing methods either rely on successive scene expansion via inpainting or employ panorama representation to represent large FOV scene environments. However, the generated scene suffers from semantic drift during expansion and is unable to handle occlusion among scene hierarchies. To tackle these challenges, we introduce LayerPano3D, a novel framework for full-view, explorable panoramic 3D scene generation from a single text prompt. Our key insight is to decompose a reference 2D panorama into multiple layers at different depth levels, where each layer reveals the unseen space from the reference views via diffusion prior. LayerPano3D comprises multiple dedicated designs: 1) we introduce a novel text-guided anchor view synthesis pipeline for high-quality, consistent panorama generation. 2) We pioneer the Layered 3D Panorama as underlying representation to manage complex scene hierarchies and lift it into 3D Gaussians to splat detailed 360-degree omnidirectional scenes with unconstrained viewing paths. Extensive experiments demonstrate that our framework generates state-of-the-art 3D panoramic scene in both full view consistency and immersive exploratory experience. We believe that LayerPano3D holds promise for advancing 3D panoramic scene creation with numerous applications.
I interviewed my daughter every first day of school since kindergarten. The last one is now done, since she is a senior. 🥹.
Here's how it turned out. https://t.co/EeoRt9Pr3l