FoundationPose

LingBot-Map is a feed-forward 3D foundation model from the Robbyant Team for streaming 3D reconstruction from image sequences and video. Its Geometric Context Transformer unifies coordinate grounding, dense geometric cues, and long-range drift correction through anchor context, a pose-reference window, and trajectory memory. The system uses paged KV-cache attention for streaming inference and supports interactive browser-based reconstruction with point-cloud visualization, keyframe intervals, windowed processing for long sequences, and sky masking. It also provides an offline rendering pipeline for image folders or video, along with evaluation pipelines for datasets including KITTI and Oxford Spires. The repository reports approximately 20 FPS inference at 518×378 resolution on sequences exceeding 10,000 frames, and documents PyTorch installation with FlashInfer as the recommended attention backend and native SDPA as a fallback.

ABot-Recon is a streaming 3D-reconstruction model and inference tool for reconstructing long video streams from video frames alone. At each step, it caches key-value features from the preceding 11 frames, predicts a point map in the current camera coordinate system, estimates the adjacent relative camera pose, and composes those poses sequentially into a global trajectory and point cloud without persistent learned long-range memory. It can output camera poses, relative poses, local and world point maps, colors, confidence maps, and run metadata. An optional loop-closure backend retrieves candidate revisited-frame pairs using DINOv2-SALAD descriptors, predicts relative-pose constraints with ABot-Recon, and refines the trajectory through sparse pose-graph optimization. The repository provides Python and command-line interfaces, a public model checkpoint, a demo, evaluation code, and visualization utilities. The released configuration targets Linux, Python 3.10 or later, PyTorch 2.5.1, and CUDA 12.1; source code is licensed under Apache License 2.0, while the model weights have separate terms in MODEL_LICENSE.md.

τ₀-VLA is an open-source hierarchical robot foundation model and reference implementation for long-horizon manipulation. A memory-augmented high-level policy generates the next subtask and uses world-model-guided test-time computation to search over alternatives when additional reasoning is needed; a generalist low-level policy then executes the selected subtask across robot embodiments. The low-level policy combines a Qwen3.5 vision-language backbone with a Mixture-of-Transformers action expert trained through conditional flow matching, using a unified 40-dimensional state/action space and multimodal co-training on tens of thousands of hours of heterogeneous real-world robot data. The repository provides LeRobot-format data loading, prompting, masking and normalization, embodiment-specific adapters, training and post-training recipes, deployment contracts, a policy server, and open-loop evaluation tools. It includes an AgiBot World example subset and templates for other datasets and robots. Public v1 serving supports joint-control checkpoints; native end-effector data can be used for training but end-effector serving is not supported in that release. The reference environment uses Python 3.11, CUDA 12.8, and PyTorch 2.7.1. Code and model weights are released under the Apache License 2.0.

Segment Anything Model (SAM) is an image-segmentation model from Meta AI Research's FAIR. It produces object masks from prompts such as points or bounding boxes, or can automatically generate masks for all objects in an image. The repository provides Python and command-line inference interfaces, pretrained checkpoints, example notebooks, and optional ONNX export and COCO-format mask processing. In the cited robotics workflow, SAM is used with Foundation Pose to extract object goal poses from a human video demonstration. The model was trained on 11 million images and 1.1 billion masks, according to the repository.

Colite is an experimental C++17 structure-from-motion and multi-view stereo system that reconstructs 3D scenes from overlapping images. Its pipeline performs incremental structure-from-motion with support for multiple models, continuation, and Sim(3) merging; feature extraction and matching; PatchMatch stereo with geometric consistency; stereo fusion; meshing; simplification; and texture-atlas generation. It produces COLMAP-compatible sparse models, RGB and normal dense point clouds, Poisson or Delaunay meshes, and textured OBJ models with MTL and atlas PNG files. CPU and optional CUDA implementations are provided for SIFT, descriptor matching, and PatchMatch Stereo, and resumable stage manifests allow processing to continue from saved stages. The repository describes the current build as alpha-quality, primarily developed and tested on Ubuntu 22.04 and WSL2; native Windows builds are not yet validated as a supported release target.