# Multi-modal On-Device Learning for Monocular Depth Estimation on Ultra-low-power MCUs

arXiv: https://arxiv.org/abs/2512.00086 · HTML: https://mlsanity.com/paper/2512.00086/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 27 of its 53 references that are in the map.
- First posted on arXiv 26 Nov 2025.

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

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Visualizing and Understanding Convolutional Networks](https://mlsanity.com/paper/1311.2901/) (2013)
- [Depth Anything V2](https://mlsanity.com/paper/2406.09414/) (2024)
- [Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data](https://mlsanity.com/paper/2401.10891/) (2024)
- [Depth Map Prediction from a Single Image using a Multi-Scale Deep Network](https://mlsanity.com/paper/1406.2283/) (2014)
- [Unsupervised Monocular Depth Estimation with Left-Right Consistency](https://mlsanity.com/paper/1609.03677/) (2016)
- [ZoeDepth: Zero-shot Transfer by Combining Relative and Metric Depth](https://mlsanity.com/paper/2302.12288/) (2023)
- [TartanAir: A Dataset to Push the Limits of Visual SLAM](https://mlsanity.com/paper/2003.14338/) (2020)
- [Metric3Dv2: A Versatile Monocular Geometric Foundation Model for Zero-shot Metric Depth and Surface Normal Estimation](https://mlsanity.com/paper/2404.15506/) (2024)
- [A Study of BFLOAT16 for Deep Learning Training](https://mlsanity.com/paper/1905.12322/) (2019)
- [Medical Image Segmentation Review: The success of U-Net](https://mlsanity.com/paper/2211.14830/) (2022)
- [On-Device Training Under 256KB Memory](https://mlsanity.com/paper/2206.15472/) (2022)
- [MonoViT: Self-Supervised Monocular Depth Estimation with a Vision Transformer](https://mlsanity.com/paper/2208.03543/) (2022)
- [TinyOL: TinyML with Online-Learning on Microcontrollers](https://mlsanity.com/paper/2103.08295/) (2021)
- [Towards real-time unsupervised monocular depth estimation on CPU](https://mlsanity.com/paper/1806.11430/) (2018)
- [On-device Training: A First Overview on Existing Systems](https://mlsanity.com/paper/2212.00824/) (2022)
- [TinyTrain: Resource-Aware Task-Adaptive Sparse Training of DNNs at the Data-Scarce Edge](https://mlsanity.com/paper/2307.09988/) (2023)
- [Human Pose Estimation on Privacy-Preserving Low-Resolution Depth Images](https://mlsanity.com/paper/2007.08340/) (2020)
- [Reduced Precision Floating-Point Optimization for Deep Neural Network On-Device Learning on MicroControllers](https://mlsanity.com/paper/2305.19167/) (2023)
- [GAP9Shield: A 150GOPS AI-capable Ultra-low Power Module for Vision and Ranging Applications on Nano-drones](https://mlsanity.com/paper/2407.13706/) (2024)
- [Self-Learning for Personalized Keyword Spotting on Ultra-Low-Power Audio Sensors](https://mlsanity.com/paper/2408.12481/) (2024)
- [Real-time Monocular Depth Estimation on Embedded Systems](https://mlsanity.com/paper/2308.10569/) (2023)
- [Benchmark data and method for real-time people counting in cluttered scenes using depth sensors](https://mlsanity.com/paper/1804.04339/) (2018)
- [Training on the Fly: On-device Self-supervised Learning aboard Nano-drones within 20 mW](https://mlsanity.com/paper/2408.03168/) (2024)
- [HyperCam: Low-Power Onboard Computer Vision for IoT Cameras](https://mlsanity.com/paper/2501.10547/) (2025)
- [A Tiny Supervised ODL Core with Auto Data Pruning for Human Activity Recognition](https://mlsanity.com/paper/2408.01283/) (2024)

## Abstract (arXiv)

Monocular depth estimation (MDE) plays a crucial role in enabling spatially-aware applications in Ultra-low-power (ULP) Internet-of-Things (IoT) platforms. However, the limited number of parameters of Deep Neural Networks for the MDE task, designed for IoT nodes, results in severe accuracy drops when the sensor data observed in the field shifts significantly from the training dataset. To address this domain shift problem, we present a multi-modal On-Device Learning (ODL) technique, deployed on an IoT device integrating a Greenwaves GAP9 MicroController Unit (MCU), a 80 mW monocular camera and a 8 x 8 pixel depth sensor, consuming ≈300mW. In its normal operation, this setup feeds a tiny 107 k-parameter μPyD-Net model with monocular images for inference. The depth sensor, usually deactivated to minimize energy consumption, is only activated alongside the camera to collect pseudo-labels when the system is placed in a new environment. Then, the fine-tuning task is performed entirely on the MCU, using the new data. To optimize our backpropagation-based on-device training, we introduce a novel memory-driven sparse update scheme, which minimizes the fine-tuning memory to 1.2 MB, 2.2x less than a full update, while preserving accuracy (i.e., only 2% and 1.5% drops on the KITTI and NYUv2 datasets). Our in-field tests demonstrate, for the first time, that ODL for MDE can be performed in 17.8 minutes on the IoT node, reducing the root mean squared error from 4.9 to 0.6m with only 3 k self-labeled samples, collected in a real-life deployment scenario.