Embodied-AI-Daily
A daily-updated collection of arXiv papers on embodied AI: vision-language navigation
MVSGaussian é um projeto de IA 3D de código aberto de TQTQliu (TQTQliu/MVSGaussian no GitHub). Tem 567 estrelas no GitHub e 37 forks, é publicado sob a licença MIT e foi atualizado pela última vez no GitHub em 25 de fevereiro de 2026.
[ECCV 2024] MVSGaussian: Quick Generalizable Gaussian Splatting Reconstruction
Clone TQTQliu/MVSGaussian and move into the project folder.
Install the Python requirements with pip — a virtual environment keeps things clean.
Gather the images or data the model expects.
Launch training or inference and inspect the 3D output.
Focused on feed forward gaussian splatting as a core part of the project.
3D scenes via Gaussian splatting.
Generalizes to new scenes.
Multi-view stereo input.
Novel view synthesis.
Python codebase — approachable for contributors.
Download the source and move into the project folder.
git clone https://github.com/TQTQliu/MVSGaussian.git
cd MVSGaussianSet up what the project needs to run.
pip install -r requirements.txtStart the app — the README has the exact command and flags.
python main.py # entrypoint varies — check the READMEMIT licensed — free for personal and commercial use.
Clone the repo and pip-install the requirements — the README covers the exact steps.
The GitHub repo's issues and discussions are the main support channel.
Last pushed 2026-02-25 with 37 forks. Skim recent commits to gauge momentum before relying on it.
Yes — it is open source, so issues and pull requests are welcome.
A daily-updated collection of arXiv papers on embodied AI: vision-language navigation
[npj AI] 3D Magic Mirror: Clothing Reconstruction from a Single Image
The official code for HUGS: Human Gaussian Splats (CVPR 2024)
Recurrent Gaussian Splatting from an ECCV spotlight paper.
ICLR 2024] Live Photorealistic Dynamic Scene Representation and Rendering.
Baking Gaussian Splatting into Diffusion Denoiser for Fast and Scalable Single-stage Image-to-3D Generation and Reconstruction (ICCV 2025)
Os outros projetos de IA 3D com mais estrelas no catálogo do LibreAI — a lista realista de alternativas se MVSGaussian não for o ideal para você.
Control Blender with natural language: an MCP plugin that lets any LLM model, light and animate in 3D
Pascal Editor: an open-source 3D architectural editor with MCP tools for humans and AI agents
Give your AI agent CAD superpowers - generate real STEP/STL models from text
Turn a reference image into a procedural, animation-ready Three.js model - as code
High-Resolution 3D Assets Generation with Large Scale Hunyuan3D Diffusion Models.
Projetos de IA 3D próximos no catálogo, os mais próximos em estrelas de MVSGaussian: continue explorando dos dois lados dele.
Bilingual AI interview cheat sheets plus a fact-checked academic homepage builder.
DiffSplat, in Python — built for [iclr 2025] official implementation of "diffsplat: repu.
A fast, friendly viewer for 3D Gaussian splat scenes.
LiDAR-augmented Gaussian Splatting plus neural SDF, in real time.
Python command-line utility designed for converting and processing 3D Gaussian.
Mesh-in-the-loop Gaussian Splatting from a SIGGRAPH Asia paper.
Grátis e aberto: pegue o código-fonte, dê uma estrela no repositório e comece a criar.