EffectPop Candy photo filter

Candy Fast Neural Style Transfer ONNX model
Source model: https://github.com/pytorch/examples/tree/main/fast_neural_style
Copyright (c) 2017, Pytorch contributors. BSD 3-Clause.
ONNX export: https://huggingface.co/onnxmodelzoo/candy-9
Publisher's model-specific license: BSD 3-Clause (model card, License section).
ONNX Model Zoo package license: Apache 2.0.
Both notices are retained in LICENSE-PyTorch-examples.txt and
LICENSE-ONNX-Model-Zoo.txt.

Revision: f24fcf558bd4564360f41cf3b5321975a33f962f
Original file: candy-9.onnx
Distributed file: candy-9-f24fcf5.onnx
SHA-256: 9d11a3529d1e547da6ae07201d93484dbab2ec0a3614535752c8f40f0fe2968a
EffectPop hosts the ONNX conversion bytes unchanged under a versioned name.
The graph runs at a fixed 224 × 224 input size. The editor fits the photo
without stretching, pads its edges, and crops the result before PNG export.
ONNX model card contributor: Jennifer Wang.

Research references:
Justin Johnson, Alexandre Alahi, Li Fei-Fei.
Perceptual Losses for Real-Time Style Transfer and Super-Resolution.
https://arxiv.org/abs/1603.08155
Dmitry Ulyanov, Andrea Vedaldi, Victor Lempitsky.
Instance Normalization: The Missing Ingredient for Fast Stylization.
https://arxiv.org/abs/1607.08022

The editor reuses ONNX Runtime Web 1.22.0 under the MIT License.
See /models/background-removal/NOTICES.txt for the runtime and dependency notices.