# EarthSight: A Distributed Framework for Low-Latency Satellite Intelligence

arXiv: https://arxiv.org/abs/2511.10834 · HTML: https://mlsanity.com/paper/2511.10834/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 16 of its 55 references that are in the map.
- First posted on arXiv 13 Nov 2025.

## Builds on (most cited first, 16 of 16)

- [Knowledge Distillation: A Survey](https://mlsanity.com/paper/2006.05525/) (2020)
- [EuroSAT: A Novel Dataset and Deep Learning Benchmark for Land Use and Land Cover Classification](https://mlsanity.com/paper/1709.00029/) (2017)
- [An Overview of Multi-Task Learning in Deep Neural Networks](https://mlsanity.com/paper/1706.05098/) (2017)
- [Once-for-All: Train One Network and Specialize it for Efficient Deployment](https://mlsanity.com/paper/1908.09791/) (2019)
- [Cross-stitch Networks for Multi-task Learning](https://mlsanity.com/paper/1604.03539/) (2016)
- [Mixture-of-Depths: Dynamically allocating compute in transformer-based language models](https://mlsanity.com/paper/2404.02258/) (2024)
- [PatternNet: A Benchmark Dataset for Performance Evaluation of Remote Sensing Image Retrieval](https://mlsanity.com/paper/1706.03424/) (2017)
- [Advancing Model Pruning via Bi-level Optimization](https://mlsanity.com/paper/2210.04092/) (2022)
- [FedSpace: An Efficient Federated Learning Framework at Satellites and Ground Stations](https://mlsanity.com/paper/2202.01267/) (2022)
- [Artificial intelligence to advance Earth observation: : A review of models, recent trends, and pathways forward](https://mlsanity.com/paper/2305.08413/) (2023)
- [Unleashing the Power of Multi-Task Learning: A Comprehensive Survey Spanning Traditional, Deep, and Pretrained Foundation Model Eras](https://mlsanity.com/paper/2404.18961/) (2024)
- [Accelerated Training for Massive Classification via Dynamic Class Selection](https://mlsanity.com/paper/1801.01687/) (2018)
- [FLuID: Mitigating Stragglers in Federated Learning using Invariant Dropout](https://mlsanity.com/paper/2307.02623/) (2023)
- [Approximation Algorithms for Stochastic Boolean Function Evaluation and Stochastic Submodular Set Cover](https://mlsanity.com/paper/1303.0726/) (2013)
- [Real-Time Aerial Fire Detection on Resource-Constrained Devices Using Knowledge Distillation](https://mlsanity.com/paper/2502.20979/) (2025)
- [Improving Satellite Imagery Masking using Multi-task and Transfer Learning](https://mlsanity.com/paper/2412.08545/) (2024)

## Built on by (most cited first, 1 of 1)

- [Exploiting answer-invariant redundancies in satellite imagery for efficient VLM inference on edge](https://mlsanity.com/paper/2609.29029/) (2026)

## Abstract (arXiv)

Low-latency delivery of satellite imagery is essential for time-critical applications such as disaster response, intelligence, and infrastructure monitoring. However, traditional pipelines rely on downlinking all captured images before analysis, introducing delays of hours to days due to restricted communication bandwidth. To address these bottlenecks, emerging systems perform onboard machine learning to prioritize which images to transmit. However, these solutions typically treat each satellite as an isolated compute node, limiting scalability and efficiency. Redundant inference across satellites and tasks further strains onboard power and compute costs, constraining mission scope and responsiveness. We present EarthSight, a distributed runtime framework that redefines satellite image intelligence as a distributed decision problem between orbit and ground. EarthSight introduces three core innovations: (1) multi-task inference on satellites using shared backbones to amortize computation across multiple vision tasks; (2) a ground-station query scheduler that aggregates user requests, predicts priorities, and assigns compute budgets to incoming imagery; and (3) dynamic filter ordering, which integrates model selectivity, accuracy, and execution cost to reject low-value images early and conserve resources. EarthSight leverages global context from ground stations and resource-aware adaptive decisions in orbit to enable constellations to perform scalable, low-latency image analysis within strict downlink bandwidth and onboard power budgets. Evaluations using a prior established satellite simulator show that EarthSight reduces average compute time per image by 1.9x and lowers 90th percentile end-to-end latency from first contact to delivery from 51 to 21 minutes compared to the state-of-the-art baseline.