🐍 Geometric Computer Vision Library for Spatial AI

InternRobotics' open platform for building generalized navigation foundation models.

Official implement of VGGT-Long

Generalizable Perception Stack for all things 3D, 4D & Scene Understanding

Real-Time Spatio-Semantic Memory

Terra: Hierarchical Terrain-Aware 3D Scene Graph for Task-Agnostic Outdoor Mapping

A Synthetic Benchmark for Evaluating Spatial Intelligence in Visual Foundation Models

Spatial analysis using Deep Spatio Temporal Point Process (DeepSTPP) updated with state-of-the-art mamba state space model, using spatiotemporal Hawkes Process, spatiotemporal self-correcting process, NJ COVID-19 and Japanese earthquake data.

Feed-forward video-to-3D scene reconstruction using VGGT (CVPR 2025) with open-vocabulary semantic labeling via SAM 2.1 + CLIP. One command, no COLMAP, Apple Silicon + CUDA.

Stochastic Preconditiong for Neural Field Optimization -- additional website assets

3D scene reconstruction from monocular video — COLMAP + 3D Gaussian Splatting + Grounded-SAM-2, with open-vocabulary CLIP queries and an agentic QA release gate.

repo to describe the floor layout project

Source-backed guide to AI world models, generated worlds, spatial AI, and model release signals

A curated benchmark zoo for embodied navigation tasks, datasets, metrics, leaderboards, and reproducibility notes.

Autonomous agent for SF design-build firms: finds AI-homeowner leads, generates 3D remodel proposals, drafts outreach, and learns from outcomes.

Reproduce Neural Assets in PyTorch with a clean, paper-faithful implementation for MoVi training and research

AI-powered indoor spatial navigation — Where Google Maps ends, Inevi begins. Live camera location detection, conversational AI guide, multilingual (EN/TE/HI), built with Groq, AWS Aurora DSQL, DynamoDB, S3, and Vercel.

Hybrid Learning-to-Rank system processing 2.4M Yelp reviews. Features a custom NLP pipeline (SBERT + VADER), Neural Ranking architecture, and MMR diversity re-ranking to solve the Cold-Start problem.

AI-driven spatial modeling, ensemble learning, and LLM-integrated decision systems for yard-scale pollinator plant recommendations

Real-time autonomous navigation system powered by monocular depth estimation and object detection. Implements Depth Anything V2 for dense spatial awareness to enable intelligent obstacle avoidance without specialized LiDAR hardware.