[arxiv 25] Turbo-VAED: Fast and Stable Transfer of Video-VAEs to Mobile Devices

🚀🚀 「大模型」2小时完全从0训练26M的小参数GPT！🌏 Train a 26M-parameter GPT from scratch in just 2h!

[3DV 2025 Best Paper] We present Object Images (Omages): An homage to the classic Geometry Images.

An open-source impl. of Large Reconstruction Models

Automated multi‑GPU/multi‑host orchestration via `SSH` (or other methods) to remotely launch parallel Wandb Sweep agents.

End-to-end Point Cloud Correspondences with Transformers

The source code of the CVPR paper "Multi-instance Point Cloud Registration by Efficient Correspondence Clustering"

[ICCV 2023] Tracking Anything with Decoupled Video Segmentation

[NeurIPS 2023] Official implementation of the paper "Segment Everything Everywhere All at Once"

[NeurIPS 2024] Unsupervised Hierarchy-Agnostic Segmentation: Parsing Semantic Image Structure

Official Implementation of paper "MonST3R: A Simple Approach for Estimating Geometry in the Presence of Motion"

A project page template for academic papers. Demo at https://eliahuhorwitz.github.io/Academic-project-page-template/

[CVPR 2023] Building Rearticulable Models for Arbitrary 3D Objects from 4D Point Clouds

A PyTorch Polynomial and Fourier series acceleration modules

[NeurIPS 2024] Official code for "Splatter a Video: Video Gaussian Representation for Versatile Processing"

Full python interactive 3D Gaussian Splatting viewer for real-time editing and analyzing.

[CVPR 2024] Official PyTorch implementation of SuGaR: Surface-Aligned Gaussian Splatting for Efficient 3D Mesh Reconstruction and High-Quality Mesh Rendering

[ECCV 2024] Reconstruction and Simulation of Elastic Objects with Spring-Mass 3D Gaussians

[CVPR 2024] 4D Gaussian Splatting for Real-Time Dynamic Scene Rendering

[CVPR 2024] REACTO: Reconstructing Articulated Objects from a Single Video

[CVPR 2024] RoMa: Robust Dense Feature Matching; RoMa is the robust dense feature matcher capable of estimating pixel-dense warps and reliable certainties for almost any image pair.

GaussianObject: High-Quality 3D Object Reconstruction from Four Views with Gaussian Splatting (SIGGRAPH Asia 2024, TOG)

[TPAMI] Gamba: Marry Gaussian Splatting with Mamba for single view 3D reconstruction

Single Image to 3D using Cross-Domain Diffusion for 3D Generation

[ICLR 2024 Oral] Generative Gaussian Splatting for Efficient 3D Content Creation

[CVPR'24] Consistent Novel View Synthesis without 3D Representation

[ICCV 2023] Single-Stage Diffusion NeRF

A markup-based typesetting system that is powerful and easy to learn.