Code implementation of Pi-Long

vision-transformer-tutorials

(ViT) Vision Image Transformer PyTorch Implementation

Linux on Dell Pro Max 18 Plus (MB18250) with Every Add-on (WWAN, SmartCard, Fingerprint etc)

An iOS app that collects/streams posed images for NeRFs using ARKit

Pillow plugin for JPEG-XL, using Rust for bindings.

Pytorch Implementation of "ActMVS: Active Scene Reconstruction with Monocular Multi-View Stereo" for ICRA 2026.

Self-hosted AI workspace.

Official implementation for paper: InFusion: Inpainting 3D Gaussians via Learning Depth Completion from Diffusion Prior

The reimplementation of megasurf

Code for CVPR 2018 paper --- Texture Mapping for 3D Reconstruction with RGB-D Sensor

[ECCV 2026] Octic Vision Transformers: Quicker ViTs Through Equivariance

[CVPR 2025 Best Paper Award] VGGT: Visual Geometry Grounded Transformer

Semantic segmentation of outdoor panoramic images using UNet-stdconv and UNet-equiconv. CVRG-Pano: semantically annotated outdoor panoramic image dataset.

Independent technical evaluations of state-of-the-art computer vision and 3D reconstruction models

GLUEMAP: Global Structure-from-Motion Meets Feedforward Reconstruction

Pop-up dictionary browser extension for language learning. Successor to Yomichan.

Official Code release for the SIGGRAPH 2026 paper "A LoD of Gaussians: Out-of-Core Training and Rendering for Seamless Ultra-Large Scene Reconstruction"

AnyCalib: On-Manifold Learning for Model-Agnostic Single-View Camera Calibration (ICCV 2025)

HorizonStream Official Implementation

control Mi Sphere 360 camera from command line

[CVPR26] MuM's a pretty good feature extractor for 3D tasks, probably the best.

Implementation of 2D-parametrization algorithm introduced by the paper "Scalable Locally Injective Mappings". Currently support only the Symmetric Dirichlet isometric energy.

Source code for the ECCV 2022 paper "Benchmarking Localization and Mapping for Augmented Reality".