An open-source project dedicated to tracking and segmenting any objects in videos, either automatically or interactively. The primary algorithms utilized include the Segment Anything Model (SAM) for key-frame segmentation and Associating Objects with Transformers (AOT) for efficient tracking and propagation purposes.

LightTrack: A Generic Framework for Online Top-Down Human Pose Tracking

RVOS: End-to-End Recurrent Network for Video Object Segmentation (CVPR 2019)

[CVPR-2024] The First High Definition (HD) Event based Visual Object Tracking Benchmark Dataset

[ECCV 2024] Elysium: Exploring Object-level Perception in Videos via MLLM

A large-scale benchmark dataset for color-event based visual tracking

A visual object tracking paper list, articles related to visual object tracking have been documented.

[WACV 2024] Separable Self and Mixed Attention Transformers for Efficient Object Tracking

Motion capture arena using only Raspberry Pi and open source libraries

Exploring through 7 popular datasets for visual object tracking, including OTB, UAV, VOT, LaSOT, NFS, TrackingNet and GOT-10k.

[BMVC 2023] Mobile Vision Transformer-based Visual Object Tracking

Modifications to improve single object tracking in 360° equirectangular videos.

A Python implementation of "Exploiting the Circulant Structure of Tracking-by-Detection with Kernels"

A complete paper list of Single-Object Visual Tracking Algorithms, Surveys and Benchmarks of recent years. Different from existing paper list, this project doesn't simply category the papers by publishment, but from a tracking challenge-tackling perspective.

MATLAB code for BACF object tracker

[ACCV 2024 (Oral, Best Application Paper)] Official Implementation of NT-VOT211: A Large-Scale Benchmark for Night-time Visual Object Tracking

[ACCV 2024 (Oral)] Official Implementation of Depth Attention for Robust RGB Tracking

Official code of "AAA: Adaptive Aggregation of Arbitrary Online Trackers with Theoretical Performance Guarantee"

Official code repository to download the TREK-150 benchmark dataset and run experiments on it.

IST-ROS: Interactive Segmentation and Tracking for ROS, is a package based on integration of Segment Anything Model with semi-supervised video object segmentation method, for flexible object segmentation and tracking in ROS-based systems.