foolwood/SiamMask

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

3547 stars · 804 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

Official PyTorch implementation of SiamMask, a deep learning framework for fast online visual object tracking and video object segmentation. It unifies tracking and segmentation in a single network, published at CVPR 2019 and extended in TPAMI 2023.

Use cases

• track objects in video in real time
• segment a target object across video frames
• reproduce CVPR 2019 object tracking results
• train a siamese tracking model from scratch
• benchmark on VOT2018 visual tracking
• generate binary masks for moving objects in video

When to choose

• you need a proven, published model for joint object tracking and segmentation
• you want official training and inference code with reproducible paper results
• you need real-time tracking performance on GPU hardware

When to avoid

• you need a maintained production library with modern PyTorch support - the code targets old PyTorch 0.4.1 and Python 3.6
• you need multi-object tracking or detection, since SiamMask tracks a single initialized target
• you want the latest state-of-the-art trackers rather than a 2019 research baseline

Facets

library · maturity maintenance

computer-vision image-processing deep-learning machine-learning computer-vision image-processing deep-learning machine-learning python object-tracking video-object-segmentation pytorch siamese-network cvpr2019 research-code video linux gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="foolwood/SiamMask")