# dwofk/fast-depth

ICRA 2019 "FastDepth: Fast Monocular Depth Estimation on Embedded Systems"

Repository: https://github.com/dwofk/fast-depth
Canonical: https://ross.abutalabs.com/products/fast-depth
Language: Python
License: MIT
License Family: permissive
Last push: 2022-12-17T07:31:45+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2753, "days_push": 1355, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1023, forks 189 (observed 2026-08-28T04:03:16.234623+00:00)

## What it is
FastDepth is the official PyTorch implementation of the ICRA 2019 paper 'FastDepth: Fast Monocular Depth Estimation on Embedded Systems' from MIT. It provides trained MobileNet-based models, evaluation code on the NYU Depth V2 dataset, and guidance for deploying real-time monocular depth estimation on embedded devices like Jetson boards.

## Use cases
- estimate depth from a single RGB image
- run real-time monocular depth estimation on embedded hardware like Jetson TX2
- evaluate depth estimation models on the NYU Depth V2 dataset
- use a lightweight pruned MobileNet depth network for robotics
- benchmark fast depth estimation networks on CUDA GPUs
- deploy a depth estimation model with TensorRT on an embedded system

## When to choose
- you need fast monocular depth estimation on resource-constrained embedded hardware
- you want pretrained lightweight depth estimation models with evaluation code
- you are doing robotics research that requires real-time depth from a single camera

## When to avoid
- you need state-of-the-art depth accuracy rather than speed
- you need stereo or multi-frame depth estimation
- you need actively maintained code on recent PyTorch versions

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, computer-vision, image-processing
- domain: computer-vision, machine-learning, robotics, embedded-systems
- platform: python, embedded
- tags: depth-estimation, monocular-depth, mobilenet, pytorch, embedded-systems, icra-2019, model-pruning, nyu-depth-v2, linux, gpu

## Member repositories
- dwofk/fast-depth (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:16.234623+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T07:08:37.744505+00:00, confidence not recorded.
  - readme: https://github.com/dwofk/fast-depth (fetched 2026-08-28T04:03:16.234623+00:00, sha ecabe76008d0)
- Data as of 2026-08-30T08:39:29.467469+00:00.