Matthias Humt

Create detailed, textured 3D meshes of anything from a short smartphone video.

Publication-ready visualization of 3D objects and point clouds in seconds.

Extract equations and context from research papers for LLM coding assistants (arXiv, LaTeX, RAG)

PyTorch implementation of 'Vanilla' Gradient, Grad-CAM, Guided backprop, Integrated Gradients and their SmoothGrad variants.

An easy-to-use wrapper around some of Open3D's registration functionality.

Registration algorithms (e.g. ICP) for Python with PCL backend.

My personal blog:

Actively maintained fork of SDFStudio: Neural implicit surface reconstruction with modern PyTorch, RTX 40XX support, PBR exports, and bug fixes

Darktide mod: smarter bots with unlocked abilities

Panel-by-panel comic reader with automatic detection, human overrides, and cross-platform PWA viewer

Code and report for the final project of the Computational Game Theory course at Vrije Universitetit Brussel (VUB) WS2017.

Source-backed Darktide entity resolution and build audit tooling. Maps community names to canonical game entities with decompiled source citations.

Minimal Python scraper/notifier for Kleinanzeigen that emails and ntfy-pushes new listings for your saved searches.

The HTML Presentation Framework

RAG on Everything with LEANN. Enjoy 97% storage savings while running a fast, accurate, and 100% private RAG application on your personal device.

Instant neural graphics primitives: lightning fast NeRF and more

A collaboration friendly studio for NeRFs

A small Python pipeline that uses gallery-dl and CLIP-based aesthetic scoring to download, curate, and filter high‑quality landscape wallpapers from selected artists’ online portfolios.

Template for explore-topic-X-from-scratch repos (Marimo + PyTorch + curated references)

VGGSfM: Visual Geometry Grounded Deep Structure From Motion

The repository provides code for running inference with the Meta Segment Anything Model 2 (SAM 2), links for downloading the trained model checkpoints, and example notebooks that show how to use the model.

Command-line interface for HLoc - deep learning-based feature extraction and matching for Structure-from-Motion

🪨 why use many token when few token do trick — Claude Code skill that cuts 65% of tokens by talking like caveman

Minimal Linux FIFO‑controlled Faster‑Whisper PTT daemon (sxhkd + systemd)

Simulation of Kinect Measurements

A procedural Blender pipeline for photorealistic training image generation

Profile README and shared GitHub Actions workflows

Easy-to-use Gradio web interface for Stable Diffusion text-to-image generation

YCB and BigBird 3D mesh reconstruction from high-res RGB images with known camera poses and intrinsics.

Python SDK, Proxy Server (AI Gateway) to call 100+ LLM APIs in OpenAI (or native) format, with cost tracking, guardrails, loadbalancing and logging. [Bedrock, Azure, OpenAI, VertexAI, Cohere, Anthropic, Sagemaker, HuggingFace, VLLM, NVIDIA NIM]

[ICCV 2023] Official code for NeuS2

No description.

This repository contains the code for the CVPR 2020 paper "Differentiable Volumetric Rendering: Learning Implicit 3D Representations without 3D Supervision"

🤗 Diffusers: State-of-the-art diffusion models for image and audio generation in PyTorch and FLAX.

Python tool for obtaining watertight meshes using TSDF fusion.

A small evolutionary simulation

3D U-Net model for volumetric semantic segmentation written in pytorch

Code for Probabilistic Shape Completion

Implicit Feature Network (IF-Net) - Codebase

[ECCV'20] Convolutional Occupancy Networks

(latest updates and bug fixed) DISN: Deep Implicit Surface Network for High-quality Single-view 3D Reconstruction

Detectron2 is FAIR's next-generation platform for object detection and segmentation.

This repository contains the code for the paper "Occupancy Networks - Learning 3D Reconstruction in Function Space"

Code and Data Release for GenRe (NeurIPS 2018) and ShapeHD (ECCV 2018)

Style-based Point Generator with Adversarial Rendering for Point Cloud Completion (CVPR 2021)

End-to-End Object Detection with Transformers

An extension of Open3D to address 3D Machine Learning tasks