# Adaptive Deep Learning for Efficient Visual Pose Estimation aboard Ultra-low-power Nano-drones

arXiv: https://arxiv.org/abs/2401.15236 · HTML: https://mlsanity.com/paper/2401.15236/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 15 of its 20 references that are in the map.
- First posted on arXiv 26 Jan 2024.

## Builds on (most cited first, 15 of 15)

- [BranchyNet: Fast Inference via Early Exiting from Deep Neural Networks](https://mlsanity.com/paper/1709.01686/) (2017)
- [DensePose: Dense Human Pose Estimation In The Wild](https://mlsanity.com/paper/1802.00434/) (2018)
- [Efficient Processing of Deep Neural Networks: A Tutorial and Survey](https://mlsanity.com/paper/1703.09039/) (2017)
- [2D/3D Pose Estimation and Action Recognition using Multitask Deep Learning](https://mlsanity.com/paper/1802.09232/) (2018)
- [DORY: Automatic End-to-End Deployment of Real-World DNNs on Low-Cost IoT MCUs](https://mlsanity.com/paper/2008.07127/) (2020)
- [Challenges and Opportunities in Edge Computing](https://mlsanity.com/paper/1609.01967/) (2016)
- [Fully Onboard AI-powered Human-Drone Pose Estimation on Ultra-low Power Autonomous Flying Nano-UAVs](https://mlsanity.com/paper/2103.10873/) (2021)
- [Neural-Swarm: Decentralized Close-Proximity Multirotor Control Using Learned Interactions](https://mlsanity.com/paper/2003.02992/) (2020)
- [Runtime Configurable Deep Neural Networks for Energy-Accuracy Trade-off](https://mlsanity.com/paper/1607.05418/) (2016)
- [An Open Source and Open Hardware Deep Learning-powered Visual Navigation Engine for Autonomous Nano-UAVs](https://mlsanity.com/paper/1905.04166/) (2019)
- [PLiNIO: A User-Friendly Library of Gradient-based Methods for Complexity-aware DNN Optimization](https://mlsanity.com/paper/2307.09488/) (2023)
- [Human Activity Recognition on Microcontrollers with Quantized and Adaptive Deep Neural Networks](https://mlsanity.com/paper/2209.00839/) (2022)
- [Learning-based Bias Correction for Time Difference of Arrival Ultra-wideband Localization of Resource-constrained Mobile Robots](https://mlsanity.com/paper/2103.01885/) (2021)
- [Deep Neural Network Architecture Search for Accurate Visual Pose Estimation aboard Nano-UAVs](https://mlsanity.com/paper/2303.01931/) (2023)
- [An IoT Endpoint System-on-Chip for Secure and Energy-Efficient Near-Sensor Analytics](https://mlsanity.com/paper/1612.05974/) (2016)

## Built on by (most cited first, 4 of 4)

- [Multi-resolution Rescored ByteTrack for Video Object Detection on Ultra-low-power Embedded Systems](https://mlsanity.com/paper/2404.11488/) (2024)
- [Optimized Deployment of Deep Neural Networks for Visual Pose Estimation on Nano-drones](https://mlsanity.com/paper/2402.15273/) (2024)
- [FiCABU: A Fisher-Based, Context-Adaptive Machine Unlearning Processor for Edge AI](https://mlsanity.com/paper/2511.05605/) (2025)
- [MR2-ByteTrack: CNN and Transformer-based Video Object Detection for AI-augmented Embedded Vision Sensor Nodes](https://mlsanity.com/paper/2605.15423/) (2026)

## Abstract (arXiv)

Sub-10cm diameter nano-drones are gaining momentum thanks to their applicability in scenarios prevented to bigger flying drones, such as in narrow environments and close to humans. However, their tiny form factor also brings their major drawback: ultra-constrained memory and processors for the onboard execution of their perception pipelines. Therefore, lightweight deep learning-based approaches are becoming increasingly popular, stressing how computational efficiency and energy-saving are paramount as they can make the difference between a fully working closed-loop system and a failing one. In this work, to maximize the exploitation of the ultra-limited resources aboard nano-drones, we present a novel adaptive deep learning-based mechanism for the efficient execution of a vision-based human pose estimation task. We leverage two State-of-the-Art (SoA) convolutional neural networks (CNNs) with different regression performance vs. computational costs trade-offs. By combining these CNNs with three novel adaptation strategies based on the output's temporal consistency and on auxiliary tasks to swap the CNN being executed proactively, we present six different systems. On a real-world dataset and the actual nano-drone hardware, our best-performing system, compared to executing only the bigger and most accurate SoA model, shows 28% latency reduction while keeping the same mean absolute error (MAE), 3% MAE reduction while being iso-latency, and the absolute peak performance, i.e., 6% better than SoA model.