# irolaina/FCRN-DepthPrediction

Deeper Depth Prediction with Fully Convolutional Residual Networks (FCRN)

Repository: https://github.com/irolaina/FCRN-DepthPrediction
Canonical: https://ross.abutalabs.com/products/fcrn-depthprediction
Language: Python
License: BSD-2-Clause
License Family: permissive
Topics: tensorflow, matconvnet, depth-maps, depth-prediction, convolutional-residual-networks
Last push: 2019-08-26T13:58:16+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 3683, "days_push": 2564, "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 1118, forks 305 (observed 2026-08-28T04:03:39.264778+00:00)

## What it is
Reference implementation and pretrained models for FCRN (Deeper Depth Prediction with Fully Convolutional Residual Networks), predicting depth maps from single RGB images. Provides TensorFlow and MatConvNet code for inference on arbitrary images and evaluation on NYU Depth v2 and Make3D benchmarks.

## Use cases
- predict depth map from a single rgb image
- estimate monocular depth with a pretrained cnn
- evaluate depth prediction on nyu depth v2
- benchmark depth estimation on make3d
- run fcrn inference in tensorflow
- get depth maps for robotics or 3d reconstruction

## When to choose
- you need the original FCRN paper models for research reproduction
- you want pretrained monocular depth estimation in TensorFlow or MatConvNet
- you need to benchmark against NYU Depth v2 or Make3D results

## When to avoid
- you need actively maintained or modern depth estimation models
- you want production-ready deployment with recent framework versions
- you need training code rather than inference
- you prefer newer architectures like MiDaS or Depth Anything

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, computer-vision, image-processing
- domain: computer-vision, machine-learning, deep-learning, robotics
- platform: python, cpp, cross-platform
- tags: depth-estimation, monocular-depth, tensorflow, matconvnet, fcrn, nyu-depth-v2, research-code, inference

## Member repositories
- irolaina/FCRN-DepthPrediction (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:39.264778+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 2026-08-30T06:41:48.384374+00:00, confidence not recorded.
  - readme: https://github.com/irolaina/FCRN-DepthPrediction (fetched 2026-08-28T04:03:39.264778+00:00, sha 0ed7b21890a9)
- Data as of 2026-08-30T08:39:29.467469+00:00.