vegidio/open-photo-ai

By vegidio · ↓ 0 · ♥ 1 · Updated 2026-03-13

What's in this repo

This page lists the files in vegidio/open-photo-ai, including fd_newyork_fp32.onnx, fr_athens_fp16.onnx, fr_athens_fp32.onnx and more.

18 ONNX models (.onnx files)
93 files, including 30 .go, 18 .onnx, 14 .xml, 12 .jpg, 6 .sum, 5 .mod
Repository size ~16.5 GB
Last updated 2026-03-13

ONNX files

File	Size	Download
models/fd_newyork_fp32.onnx	—	Download
models/fr_athens_fp16.onnx	180 MB	Download
models/fr_athens_fp32.onnx	359 MB	Download
models/fr_santorini_fp16.onnx	161 MB	Download
models/fr_santorini_fp32.onnx	322 MB	Download
models/la_paris_fp16.onnx	146 KB	Download
models/la_paris_fp32.onnx	280 KB	Download
models/up_kyoto_2x_fp16.onnx	32.2 MB	Download
models/up_kyoto_2x_fp32.onnx	64.0 MB	Download
models/up_kyoto_4x_fp16.onnx	32.1 MB	Download

Showing 10 of 18 .onnx files — see the full file list on the right.

How to use

These ONNX files can be loaded with ONNX Runtime. To inspect a model graph, upload the file to the Netron viewer. Download links point to the original files on Hugging Face.

README

Open Photo AI Models

This repository contains a collection of AI models used by Open Photo AI for photo enhancement tasks, including face detection, face recovery, image upscaling, denoising, sharpening, light adjustment, and color balance. Each model is designed to perform a specific function to improve the quality of photographs.

Models

Face Detection

New York: based on RetinaFace (ResNet34), with fixed shape of 640x640, FP32 precision.

Face Recovery

Athens: based on CodeFormer, with fixed shape of 512x512, FP32 and FP16 precisions.
Santorini: based on GFPGAN, with fixed shape of 512x512, FP32 and FP16 precisions.

Upscale

Tokyo: based on SwinIR, with fixed shape of 256x256, FP32 and FP16 precisions.
Kyoto: based on Real-ESRGAN (General), with fixed shape of 256x256, FP32 and FP16 precisions.
Saitama: based on Real-ESRGAN (Anime), with fixed shape of 256x256, FP32 and FP16 precisions.

Light Adjustment

Paris: based on IAT with dynamic shapes, FP32 and FP16 precisions.

Color Balance

Rio: coming soon...

Denoise

Stockholm: coming soon...

Sharpen

Moscow: coming soon...

Model info

onnx region:us

↓ 0 ♥ 1

Created: 2025-11-11

Updated: 2026-03-13

View on Hugging Face

Files (93)

.gitattributes

.gitignore

.idea/.gitignore

.idea/go.imports.xml

.idea/hugging-face.iml

.idea/misc.xml

.idea/modules.xml

.idea/runConfigurations/Run_Athens.xml

.idea/runConfigurations/Run_Kyoto.xml

.idea/runConfigurations/Run_New_York.xml

.idea/runConfigurations/Run_Paris.xml

.idea/runConfigurations/Run_Santorini.xml

.idea/runConfigurations/Test_Athens.xml

.idea/runConfigurations/Test_Kyoto.xml

.idea/runConfigurations/Test_New_York.xml

.idea/runConfigurations/Test_Santorini.xml

.idea/runConfigurations/Test_Tokyo.xml

.idea/vcs.xml

README.md

Taskfile.yml

facedetection/anchors.go

facedetection/decode.go

facedetection/filter.go

facedetection/get_faces.go

facedetection/go.mod

facedetection/go.sum

facedetection/input.go

facedetection/newyork/inference.go

facedetection/newyork/inference_test.go

facedetection/types.go

facerecovery/athens/face1_reference.jpg

facerecovery/athens/face2_reference.jpg

facerecovery/athens/face3_reference.jpg

facerecovery/athens/inference.go

facerecovery/athens/inference_test.go

facerecovery/face.go

facerecovery/go.mod

facerecovery/go.sum

facerecovery/inference.go

facerecovery/santorini/face1_reference.jpg

facerecovery/santorini/face2_reference.jpg

facerecovery/santorini/face3_reference.jpg

facerecovery/santorini/inference.go

facerecovery/santorini/inference_test.go

facerecovery/transform.go

facerecovery/types.go

go.work

go.work.sum

lightadjustment/blend.go

lightadjustment/go.mod

lightadjustment/go.sum

lightadjustment/paris/inference.go

models/fd_newyork_fp32.onnx ONNX

models/fr_athens_fp16.onnx 180 MB ONNX

models/fr_athens_fp32.onnx 359 MB ONNX

models/fr_santorini_fp16.onnx 161 MB ONNX

models/fr_santorini_fp32.onnx 322 MB ONNX

models/la_paris_fp16.onnx 146 KB ONNX

models/la_paris_fp32.onnx 280 KB ONNX

models/up_kyoto_2x_fp16.onnx 32.2 MB ONNX

models/up_kyoto_2x_fp32.onnx 64.0 MB ONNX

models/up_kyoto_4x_fp16.onnx 32.1 MB ONNX

models/up_kyoto_4x_fp32.onnx 64.0 MB ONNX

models/up_osaka_fp16.onnx 2.5 MB ONNX

models/up_osaka_fp16.onnx.data

models/up_osaka_vae_decoder_fp16.onnx 99.1 MB ONNX

models/up_osaka_vae_encoder_fp16.onnx 65.4 MB ONNX

models/up_saitama_4x_fp16.onnx 8.6 MB ONNX

models/up_saitama_4x_fp32.onnx 17.1 MB ONNX

models/up_tokyo_4x_fp16.onnx 65.6 MB ONNX

models/up_tokyo_4x_fp32.onnx 130 MB ONNX

shared/go.mod

shared/go.sum

shared/image.go

shared/math.go

shared/session.go

shared/tensor.go

testdata/face1_original.jpg

testdata/face1_small.jpg

testdata/face2_original.jpg

testdata/face2_small.jpg

testdata/face3_original.jpg

testdata/face3_small.jpg

upscale/blend.go

upscale/go.mod

upscale/go.sum

upscale/inference.go

upscale/kyoto/inference.go

upscale/kyoto/inference_test.go

upscale/process.go

upscale/tiling.go

upscale/tokyo/inference.go

upscale/tokyo/inference_test.go

Related models

More from vegidio

No other models by this author.

More models

No task type declared.