# SatSAM2: Motion-Constrained Video Object Tracking in Satellite Imagery using Promptable SAM2 and Kalman Priors

arXiv: https://arxiv.org/abs/2511.18264 · HTML: https://mlsanity.com/paper/2511.18264/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 27 of its 55 references that are in the map.
- First posted on arXiv 23 Nov 2025.

## Builds on (most cited first, 27 of 27)

- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [SAM 2: Segment Anything in Images and Videos](https://mlsanity.com/paper/2408.00714/) (2024)
- [MOT20: A benchmark for multi object tracking in crowded scenes](https://mlsanity.com/paper/2003.09003/) (2020)
- [Learning Discriminative Model Prediction for Tracking](https://mlsanity.com/paper/1904.07220/) (2019)
- [Hiera: A Hierarchical Vision Transformer without the Bells-and-Whistles](https://mlsanity.com/paper/2306.00989/) (2023)
- [Track Anything: Segment Anything Meets Videos](https://mlsanity.com/paper/2304.11968/) (2023)
- [Probabilistic Regression for Visual Tracking](https://mlsanity.com/paper/2003.12565/) (2020)
- [MatrixCity: A Large-scale City Dataset for City-scale Neural Rendering and Beyond](https://mlsanity.com/paper/2309.16553/) (2023)
- [MOTChallenge: A Benchmark for Single-Camera Multiple Target Tracking](https://mlsanity.com/paper/2010.07548/) (2020)
- [Learning Target Candidate Association to Keep Track of What Not to Track](https://mlsanity.com/paper/2103.16556/) (2021)
- [UIU-Net: U-Net in U-Net for Infrared Small Object Detection](https://mlsanity.com/paper/2212.00968/) (2022)
- [Know Your Surroundings: Exploiting Scene Information for Object Tracking](https://mlsanity.com/paper/2003.11014/) (2020)
- [Transforming Model Prediction for Tracking](https://mlsanity.com/paper/2203.11192/) (2022)
- [SAMURAI: Adapting Segment Anything Model for Zero-Shot Visual Tracking with Motion-Aware Memory](https://mlsanity.com/paper/2411.11922/) (2024)
- [Autoregressive Queries for Adaptive Tracking with Spatio-TemporalTransformers](https://mlsanity.com/paper/2403.10574/) (2024)
- [Tracking Meets LoRA: Faster Training, Larger Model, Stronger Performance](https://mlsanity.com/paper/2403.05231/) (2024)
- [Robust Visual Tracking by Segmentation](https://mlsanity.com/paper/2203.11191/) (2022)
- [PointSAM: Pointly-Supervised Segment Anything Model for Remote Sensing Images](https://mlsanity.com/paper/2409.13401/) (2024)
- [OmniCity: Omnipotent City Understanding with Multi-level and Multi-view Images](https://mlsanity.com/paper/2208.00928/) (2022)
- [Detecting and Tracking Small and Dense Moving Objects in Satellite Videos: A Benchmark](https://mlsanity.com/paper/2111.12960/) (2021)
- [SG-BEV: Satellite-Guided BEV Fusion for Cross-View Semantic Segmentation](https://mlsanity.com/paper/2404.02638/) (2024)
- [Enhancing Perception of Key Changes in Remote Sensing Image Change Captioning](https://mlsanity.com/paper/2409.12612/) (2024)
- [ESOD: Efficient Small Object Detection on High-Resolution Images](https://mlsanity.com/paper/2407.16424/) (2024)
- [ROS-SAM: High-Quality Interactive Segmentation for Remote Sensing Moving Object](https://mlsanity.com/paper/2503.12006/) (2025)
- [3D Building Reconstruction from Monocular Remote Sensing Images with Multi-level Supervisions](https://mlsanity.com/paper/2404.04823/) (2024)
- [Siamese-DETR for Generic Multi-Object Tracking](https://mlsanity.com/paper/2310.17875/) (2023)

## Abstract (arXiv)

Existing satellite video tracking methods often struggle with generalization, requiring scenario-specific training to achieve satisfactory performance, and are prone to track loss in the presence of occlusion. To address these challenges, we propose SatSAM2, a zero-shot satellite video tracker built on SAM2, designed to adapt foundation models to the remote sensing domain. SatSAM2 introduces two core modules: a Kalman Filter-based Constrained Motion Module (KFCMM) to exploit temporal motion cues and suppress drift, and a Motion-Constrained State Machine (MCSM) to regulate tracking states based on motion dynamics and reliability. To support large-scale evaluation, we propose MatrixCity Video Object Tracking (MVOT), a synthetic benchmark containing 1,500+ sequences and 157K annotated frames with diverse viewpoints, illumination, and occlusion conditions. Extensive experiments on two satellite tracking benchmarks and MVOT show that SatSAM2 outperforms both traditional and foundation model-based trackers, including SAM2 and its variants. Notably, on the OOTB dataset, SatSAM2 achieves a 5.84% AUC improvement over state-of-the-art methods. Our code and dataset will be publicly released to encourage further research.