google-deepmind/tapnet
Tracking Any Point (TAP)
This project helps you accurately track specific points or objects through video footage, even if they move, are temporarily hidden, or change appearance. You input a video and select a point, and it outputs the exact path (trajectory) of that point across all frames. It's ideal for robotics engineers, motion analysts, or researchers studying object movement.
1,814 stars. Actively maintained with 3 commits in the last 30 days.
Use this if you need precise, long-term tracking of individual points or features within complex videos, including scenarios with occlusions or re-detections.
Not ideal if you only need general object detection or classification rather than detailed point-level trajectory analysis.
Stars
1,814
Forks
176
Language
Jupyter Notebook
License
Apache-2.0
Category
Last pushed
Jan 22, 2026
Commits (30d)
3
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/google-deepmind/tapnet"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related tools
roboflow/rf-detr
[ICLR 2026] RF-DETR is a real-time object detection and segmentation model architecture...
stereolabs/zed-sdk
⚡️The spatial perception framework for rapidly building smart robots and spaces
mikel-brostrom/boxmot
BoxMOT: Pluggable SOTA multi-object tracking modules with support for axis-aligned and oriented...
RizwanMunawar/yolov7-object-tracking
YOLOv7 Object Tracking Using PyTorch, OpenCV and Sort Tracking
NVIDIA-ISAAC-ROS/isaac_ros_object_detection
NVIDIA-accelerated, deep learned model support for image space object detection