# baowenbo/DAIN

Depth-Aware Video Frame Interpolation (CVPR 2019)

Repository: https://github.com/baowenbo/DAIN
Canonical: https://ross.abutalabs.com/products/dain
Homepage: https://sites.google.com/view/wenbobao/dain
Language: Python
License: MIT
License Family: permissive
Last push: 2023-02-13T12:40:12+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2722, "days_push": 1297, "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 8316, forks 834 (observed 2026-08-28T04:10:20.270106+00:00)

## What it is
DAIN is a PyTorch-based deep learning model for depth-aware video frame interpolation, synthesizing intermediate frames between existing ones while handling occlusion via depth cues. It is the official research code from a CVPR 2019 paper, including pre-trained models and training scripts.

## Use cases
- interpolate missing frames in a video
- convert video to slow motion
- increase video frame rate with AI
- smooth choppy low-fps footage
- research video frame interpolation models
- generate in-between frames handling occlusion

## When to choose
- you need high-quality frame interpolation that handles occlusion and large motion
- you want a research-grade model with pre-trained weights and training code
- you have a GPU and want to generate slow-motion or fps-boosted videos

## When to avoid
- you need a fast, actively maintained tool with modern GPU support
- you want real-time interpolation or a simple GUI application
- your project requires recent deep learning framework versions out of the box

## Facets
- artifact type: library
- maturity: maintenance
- function: video-processing, machine-learning, image-processing
- domain: computer-vision, deep-learning, media
- platform: python
- tags: video-frame-interpolation, optical-flow, depth-estimation, slow-motion, cvpr-2019, research-code, video, linux, gpu

## Member repositories
- baowenbo/DAIN (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:20.270106+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 2026-08-29T17:28:03.484983+00:00, confidence not recorded.
  - readme: https://github.com/baowenbo/DAIN (fetched 2026-08-28T04:10:20.270106+00:00, sha 1093a3a3ac5c)
  - homepage: https://sites.google.com/view/wenbobao/dain (fetched 2026-08-29T08:27:43.834845+00:00, sha 3050d8bb0ed8)
- Data as of 2026-08-30T08:39:29.467469+00:00.