edge_compute

A compilation of resources for satellite edge computing. For now, this includes onboard ML and ML model distillation.

1. Onboard Processing

1.1 PyRaws for processing raw Sentinel-2 satellite imagery (imagery level 0) onboard the satellite.

Github: https://github.com/ESA-PhiLab/PyRawS
Paper: https://arxiv.org/abs/2305.11891
podcast: https://www.youtube.com/watch?v=aLe7Pw3DVmo&feature=youtu.be

1.2 Machine Learning Onboard Satellites

citation: SmartSat 2021, Machine Learning Onboard Satellites, SmartSat Technical Report no. 010, SmartSat,

Adelaide, Australia.

Report: https://smartsatcrc.lbcdn.io/uploads/Technical-report-machine-learning-onboard-satellites-PUB-1.pdf

Rating: TBD

1.3 Benchmarking Deep Learning for On-Board Space Applications

Authors: Maciej Ziaja et al
paper: https://www.mdpi.com/2072-4292/13/19/3981
Rating: TBD

2. Distilling Machine Learning Models

2.1 Distilling the Knowledge in a Neural Network

Authors: Geoffrey Hinton, Oriol Vinyals, Jeff Dean
paper: https://arxiv.org/abs/1503.02531
Rating: 5/5.

2.2 Knowledge Distillation: Principles & Algorithms [+Applications]

Author: Rohit Kundu, July 2022
Link: https://www.v7labs.com/blog/knowledge-distillation-guide
Rating: TBD

2.3 Knowledge Distillation: Principles, Algorithms, Applications

Author: Sundeep Teki, Sept 2023
Link: https://neptune.ai/blog/knowledge-distillation
Rating: TBD

2.4 Knowledge Distillation: A Survey

Authors: Jiangping Gou et al
paper: https://arxiv.org/abs/2006.05525
Rating: TBD

3. Courses/Learning resources:

Edge impulse Computer Vision with Embedded Machine Learning:

https://www.coursera.org/learn/computer-vision-with-embedded-machine-learning