Simulating Large-Scale Multi-Agent Interactions with Limited Multimodal Senses and Physical Needs

Seen2Scene takes an incomplete real-world 3D scan and generates a complete, coherent 3D scene using visibility-guided flow matching — trained directly on real-world data.

Open Brush is the open source, community led evolution of Tilt Brush! Forked from https://github.com/googlevr/tilt-brush

DiffSoup: Direct Differentiable Rasterization of Triangle Soup for Extreme Radiance Field Simplification (CVPR 2026)

Direct solver for sparse SPD matrices for nonlinear optimization. Implements supernodal Cholesky decomposition algorithm, and supports GPU (CUDA).

Visual-inertial optimization of VIO trajectories and SLAM maps via accurate sensor modelling, with imu pre-integrated terms, full re-estimation of intrinsics/extrinsics on a generic rig and modelli…

[CVPR 2026] InfiniDepth: Arbitrary-Resolution and Fine-Grained Depth Estimation with Neural Implicit Fields

Fast mesh to 3D gaussian splat conversion

[CVPR 2024 Highlight] XCube: Large-Scale 3D Generative Modeling using Sparse Voxel Hierarchies

[IEEE RA-L] Air-IO: a learning-based IO framework targeted for UAV

Augmented Vertex Block Descent (AVBD) reference implementation

Reimplementation of LoGeR: Long-Context Geometric Reconstruction with Hybrid Memory

[CVPR 2026] ZipMap: Linear-Time Stateful 3D Reconstruction via Test-Time Training

Application for camera and sensor data logging (iOS)

A ray tracing program

Ghostty-based macOS terminal with vertical tabs and notifications for AI coding agents

Bundle Adjustment in the Eager Mode

A curated survey and paper list for global solvers in 3D vision: certifiable optimization, convex relaxation, Branch-and-Bound, and robust estimation.

Real-time 3D visualization of space.

[CVPR 2026] NeoVerse: Enhancing 4D World Model with in-the-wild Monocular Videos

[CVPR'24 Oral] Official repository of Point Transformer V3 (PTv3)

Pointcept: Perceive the world with sparse points, a codebase for point cloud perception research. Latest works: Utonia, Concerto (NeurIPS'25), Sonata (CVPR'25 Highlight), PTv3 (CVPR'24 Oral)

Official repo for paper "Structured 3D Latents for Scalable and Versatile 3D Generation" (CVPR'25 Spotlight).

PLVS is a real-time SLAM system with points, lines, volumetric mapping and 3D unsupervised incremental segmentation.

"Paper2Slides: From Paper to Presentation in One Click"

(ICLR2026) ReconViaGen: Towards Accurate Multi-view 3D Object Reconstruction via Generation

CUDA Templates and Python DSLs for High-Performance Linear Algebra

Code for the ShapeR research paper

[CVPR'26] Official implementation of "C3G: Learning Compact 3D Representations with 2K Gaussians"