DetFill-with-DHT: Deterministic Hints and Hint-AUC for Line Art Colorization

Official implementation of "Hint-AUC: Deterministic Region-based Hint Generation for Line Art Colorization Evaluation", IEEE Transactions on Visualization and Computer Graphics (TVCG), 2026.

Koki Madono · Yuan Mingcheng · Edgar Simo-Serra — Waseda University

Paper · Library on PyPI · Checkpoints and data

Hint-based line art colorization is usually evaluated with randomly sampled colour hints, so the scores change from run to run. This repository provides DHT, a deterministic hint generator that turns an illustration into the same scribble and dot hints every time; Hint-AUC, one score over the whole range of hint ratios, from no hints to fully hinted; and DetFill, a pixel-space diffusion colorization model trained with these hints. All checkpoints, the evaluation inputs and the scripts that reproduce the paper are released.

Getting started

• Installation — see Setup. The library is one command (pip install hintauc); DetFill and the reproduction scripts use a conda environment that the scripts create for you.
• Model zoo — see Model zoo: every released checkpoint and data file with its size, SHA-256, download link and the paper item it belongs to.
• Reproducing the paper — see Reproducing the paper: Fig. 9 with one command (the Replicability-Stamp script), every table rebuilt from the released metrics, every experiment re-run on example images, full table rows, and training.
• Added features (since the paper) for improving our library — see Added features.
• How it works — see Detailed explanation: each component, what is bit-exact and what is not.

Quick start

import hintauc

hints = hintauc.generate_hints("illustration.png", size=64)        # image -> deterministic hints
color, mask = hints.at_ratio(0.10, hint_type="scribble")           # hints of the largest 10 % of the regions
hints.save("out/illustration")                                     # files in the DetFill data layout

evaluator = hintauc.Evaluator(metrics=("psnr", "lpips", "dreamsim"))
print(evaluator("colorized.png", "ground_truth.png"))              # the paper's metrics

• Command line: hintauc generate image.png --ratio 0.1 and hintauc eval pred_dir gt_dir --metrics mse psnr ssim.
• DetFill inference over the hint-ratio grid: cd detfill && GPU=0 bash run_inference.sh scribble (data layout and options in detfill/README.md).
• Fig. 9 of the paper from a fresh checkout: bash replicability/run.sh (about 3 minutes on a GPU).

Repository structure

docs/              setup, added features, detailed explanation
replicability/     one-command reproduction of Fig. 9 (Replicability-Stamp entry point)
reproduce/         reproduction package: scripts/, examples/, expected/ (released metrics), data/, paper_experiments/
hintauc/           the library: hints.py, metrics.py, auc.py, cli.py
detfill/           the DetFill model (BBDM fork): training, inference, configs
hint_generation/   original research scripts behind the library
evaluation/        original evaluation scripts
checkpoints/       model zoo: inventory of all released files with hashes
assets/            images for this page; representative image for the Replicability Stamp

License

MIT License (LICENSE); third-party components are listed in THIRD_PARTY_NOTICES.md. detfill/ is built on BBDM (MIT). Released data files are derived from Danbooru2021 illustrations and are provided for non-commercial research use only.

Contributing

Contributions are welcome. See contributing and the code of conduct.

Citing

@article{madono2026hintauc,
  title   = {Hint-AUC: Deterministic Region-based Hint Generation for Line Art Colorization Evaluation},
  author  = {Madono, Koki and Mingcheng, Yuan and Simo-Serra, Edgar},
  journal = {IEEE Transactions on Visualization and Computer Graphics},
  year    = {2026},
  doi     = {10.1109/TVCG.2026.3738401}
}