henry123-boy/SpaTrackerV2

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

1005 stars · 55 forks observed · 2026-10-21

What it is AI-extracted, prompt v1, taxonomy v1, 2026-09-07, confidence 0.9

SpatialTrackerV2 is a research library from ICCV 2025 that performs 3D point tracking in videos, lifting tracked points into 3D using depth models. It ships pretrained checkpoints, a Hugging Face demo, and Python inference code.

Use cases

• track 3D points in a video
• estimate camera pose and depth jointly from video
• reconstruct dynamic scenes from monocular video
• analyze object motion in 3D from RGB footage
• run 3D tracking inference on a GPU

When to choose

• you need state-of-the-art 3D point tracking for research or prototyping
• you want a pretrained model with a quick-start offline demo
• you can run inference on a CUDA GPU

When to avoid

• you need production-grade online streaming tracking (still unreleased)
• you need training or evaluation code (not yet published)
• you need a lightweight CPU-only solution

Facets

library · maturity active

machine-learning computer-vision image-processing computer-vision machine-learning deep-learning python 3d-point-tracking video depth-estimation research-code iccv-2025 gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="henry123-boy/SpaTrackerV2")