spatial-intelligence

SpatialLM: Training Large Language Models for Structured Indoor Modeling

[ICCV 2025] Aether: Geometric-Aware Unified World Modeling

Official implementation of Spatial-MLLM: Boosting MLLM Capabilities in Visual-based Spatial Intelligence

[ICLR 2025] SPA: 3D Spatial-Awareness Enables Effective Embodied Representation

[CVPR 2025] Source codes for the paper "3D-Mem: 3D Scene Memory for Embodied Exploration and Reasoning"

[NeurIPS 2024] Official code for HourVideo: 1-Hour Video Language Understanding

Machine Mental Imagery: Empower Multimodal Reasoning with Latent Visual Tokens (arXiv 2025)

[CVPR 2025] Code for "StarGen: A Spatiotemporal Autoregression Framework with Video Diffusion Model for Scalable and Controllable Scene Generation".

Source codes for the paper "MindJourney: Test-Time Scaling with World Models for Spatial Reasoning"

Multimodal datasets for spatial intelligence

[ICCV 2025] Perspective-Invariant 3D Object Detection

This is the official implementation of "LiDARCrafter: Dynamic 4D World Modeling from LiDAR Sequences"

"Gradio" Interface for SpatialLM Model | A 3D Large Language Model for Structured Scene Understanding, Processing Point Cloud Data from Monocular Videos, RGBD Images, and LiDAR.

SpatialFusion-LM is a real-time spatial reasoning framework that combines neural depth, 3D reconstruction, and language-driven scene understanding.

3D and 4D World Modeling: A Survey

SolidGeo: Measuring Multimodal Spatial Math Reasoning in Solid Geometry

Benchmarking 3D and 4D World Models in the Real World

Trying out SpatialLM (SpatialLM: Large Language Model for Spatial Understanding). Impressed with results 💖