# vt-vl-lab/3d-photo-inpainting

[CVPR 2020] 3D Photography using Context-aware Layered Depth Inpainting

Repository: https://github.com/vt-vl-lab/3d-photo-inpainting
Canonical: https://ross.abutalabs.com/products/3d-photo-inpainting
Homepage: https://shihmengli.github.io/3D-Photo-Inpainting/
Language: Python
License: NOASSERTION
License Family: other
Topics: novel-view-synthesis, 3d-photo
Last push: 2024-08-30T23:48:56+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": 2338, "days_push": 733, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 7093, forks 1132 (observed 2026-08-28T04:09:55.876491+00:00)

## What it is
A Python research codebase from a CVPR 2020 paper that converts a single RGB-D image into a 3D photo using layered depth inpainting. It hallucinates color and depth in occluded regions to enable novel view synthesis with motion parallax effects.

## Use cases
- turn a single photo into a 3D parallax animation
- generate novel views from one RGB-D image
- inpaint occluded regions of a depth image
- create dolly zoom effects from still photos
- animate old family photos with 2.5D motion
- estimate depth maps with MiDaS and render 3D photos

## When to choose
- you want to reproduce CVPR 2020 layered depth inpainting results
- you need novel view synthesis from a single image on Linux with PyTorch
- you want research-grade 3D photo effects like parallax and dolly zoom

## When to avoid
- you need a production-ready or actively maintained pipeline
- you work on Windows or macOS without Linux compatibility effort
- you need real-time or web-based 3D photo generation
- you require a permissive license for commercial use

## Facets
- artifact type: library
- maturity: maintenance
- function: image-processing, computer-vision, machine-learning, graphics, deep-learning
- domain: computer-vision, image-processing, graphics, artificial-intelligence
- platform: python
- tags: 3d-photo, novel-view-synthesis, depth-inpainting, rgb-d, parallax-effect, cvpr-2020, research-code, linux, gpu

## Member repositories
- vt-vl-lab/3d-photo-inpainting (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:55.876491+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 2026-08-29T17:40:09.698549+00:00, confidence not recorded.
  - readme: https://github.com/vt-vl-lab/3d-photo-inpainting (fetched 2026-08-28T04:09:55.876491+00:00, sha 7f35ec0d74dc)
  - homepage: https://shihmengli.github.io/3D-Photo-Inpainting/ (fetched 2026-08-29T08:35:48.080372+00:00, sha ddba17482d1d)
- Data as of 2026-08-30T08:39:29.467469+00:00.