克隆/下载
贡献代码
同步代码
取消
提示: 由于 Git 不支持空文件夾,创建文件夹后会生成空的 .keep 文件
Loading...
README
MulanPSL-2.0

基于深度学习的2D虚拟试衣

项目介绍

本项目主要面向第 $14$ 届全国服务外包创新创业比赛 $A16$ 赛道虚拟试衣赛题,采用 $2D$ 虚拟试衣技术依托于 $VITON$ 开源数据集训练 $DNN$ 网络并着重进行工程化落地应用;项目选用了前沿顶刊论文的 $PFAFN$ 模型,在此基础上对模型进行优化改进,实现了模型压缩和推理加速并使用 $OpenVINO$ 框架进行部署应用,出色地完成了赛题的要求。

项目示例

项目开发环境

开发平台 版本 开发工具 版本
Pycharm 2022.3.2 Visual Studio Code 1.80.1
Visual Studio 17.5.5
开发环境 版本 开发环境 版本
neural-compressor 2.2.1 nncf 2.5.0
numpy 1.23.4 onnx 1.14.0
opencv-python 4.7.0.72 onnxruntime 1.15.1
openvino 2022.3.0 pandas 1.3.5
pytorch-fid 0.3.0 rembg 2.0.50
pytorch 2.0.0 torch-pruning 1.1.9
intel-openmp 2021.4.0

环境配置

  • 克隆仓库
git clone https://github.com/LZHMS/Virtual-Tryon.git
  • 安装依赖库
pip install -r requirements.txt

项目文件介绍

本项目主要分为模型训练和工程化落地两部分,因此仓库创建了两个分支 mainPruingQuantization

  • main 分支是模型的推理部分,其中包括原始 Pytorch 模型、ONNX 模型、剪枝后模型、量化后模型的推理;
    • Img2Col模块用来对 corr_pure_torch模块做推理加速,模型训练中采用 corr_pure_torch模块而在推理阶段采用 Img2Col模块;
    • $afwm$ 与 $networks$ 分别是 $PFAFN$ 模型的衣服形变模块和图像生成模块
  • PruningQuantization 分支是模型工程化落地部分,其中还包括模型训练部分和模型剪枝量化;
    • ModelTraining 是 $PFAFN$ 模型的训练部分,分为四个阶段,先对教师网络进行训练然后采用可调节的知识蒸馏训练学生网络;
    • ModelPruningQuantization 是本项目主要的工程化落地部分,模型剪枝主要针对于 $Warp$ 模型,为降低模型精度损失我们采用模块化剪枝的策略,并加入模型微调,将模型分为若干模块分别进行剪枝;模型量化采用了多种量化技术以及多种量化工具,具体尝试了 $Nerual\ Compressor$ 的训练后静态量化、Pytorch 的训练后静态量化以及 Pytorch 的量化感知训练。

模型结构介绍

本项目基于 $PFAFN$ 模型重新设计各个网络模块,具体结构如下图所示:

DNN网络结构

项目工程化落地

为了满足赛题方的要求,本项目开展了工程化落地部分,主要分为两个部分,模型训练和模型剪枝量化。项目工程化部署总图如下所示: 项目工程化部署总图

实验结果:通道剪枝

  • Clothe Warp Module
Metrics GFLOPs Para(M) SIZE(MB) Total SIZE(MB) Compresion Ratio FID FID Loss
Original Module 6.63 9.37 35.8 112.0 100.00% 8.906 0.00%
Ratio=0.2 with FineTuning 5.23 7.28 27.6 88.69 79.19% 9.013 1.20%
Ratio=0.3 with FineTuning 4.40 6.48 24.8 65.73 58.69% 9.113 2.32%
Ratio=0.4 with FineTuning 3.79 5.61 20.4 40.97 36.58% 9.304 4.47%
Ratio=0.5 with FineTuning 3.42 4.55 16.8 35.47 31.67% 9.977 12.03%
  • Image Generation Module
Metrics GFLOPs Para(M) SIZE(MB) Total SIZE(MB) Compresion Ratio FID FID Loss
Original Module 21.93 43.90 167 167 100.00% 8.906 0.00%
Ratio=0.2 with FineTuning 16.54 35.02 112.3 112.3 67.25% 9.212 3.44%
Ratio=0.25 with FineTuning 15.45 31.93 94.39 94.39 56.52% 9.405 5.60%
Ratio=0.3 with FineTuning 13.90 29.89 80.25 80.25 48.05% 9.679 8.68%
Ratio=0.35 with FineTuning 12.78 27.31 73.49 73.49 44.01% 9.835 10.43%
Ratio=0.4 with FineTuning 11.20 26.12 68.52 68.52 41.03% 10.527 18.20%
  • 最优剪枝方案
Model Original Model Sparsity Pruned Model FID FPS
CWM 112MB 40% 40.97MB 9.504 2.92
IGM 167MB 25% 94.39MB 9.504 2.92

实验结果:量化感知训练

Optimization CPU-FID GPU-FID Original Model Quantized Model
Unquantized 9.504 9.483 135.36MB 135.36MB
Quantize CWM 9.783 9.701 40.97MB 10.85MB
Quantize IGM 10.382 10.249 94.39MB 24.10MB
Quantize CWM & IGM 11.503 11.379 135.36MB 34.95MB

实验结果:img2col 优化加速

Runtimes CorrTorch(s) Img2Col(s) FPS Acceleration Rate
n=1000 147.8491 94.7902 10.81 1.5598
n=10000 1489.1325 927.4293 10.77 1.6057
Average Time 0.1488 0.029 10.79 1.6017

参考文献

  • Y. Ge, Y. Song, R. Zhang, C. Ge, W. Liu, and P. Luo, "Parser-Free Virtual Try-on via Distilling Appearance Flows," arXiv preprint arXiv:2103.04559, 2021.
  • Y. Cheng, D. Wang, P. Zhou and T. Zhang, "Model Compression and Acceleration for Deep Neural Networks: The Principles, Progress, and Challenges," in IEEE Signal Processing Magazine, vol. 35, no. 1, pp. 126-136, Jan. 2018, doi: 10.1109/MSP.2017.2765695.
  • PyTorch Quantization Aware Training
木兰宽松许可证, 第2版 木兰宽松许可证, 第2版 2020年1月 http://license.coscl.org.cn/MulanPSL2 您对“软件”的复制、使用、修改及分发受木兰宽松许可证,第2版(“本许可证”)的如下条款的约束: 0. 定义 “软件”是指由“贡献”构成的许可在“本许可证”下的程序和相关文档的集合。 “贡献”是指由任一“贡献者”许可在“本许可证”下的受版权法保护的作品。 “贡献者”是指将受版权法保护的作品许可在“本许可证”下的自然人或“法人实体”。 “法人实体”是指提交贡献的机构及其“关联实体”。 “关联实体”是指,对“本许可证”下的行为方而言,控制、受控制或与其共同受控制的机构,此处的控制是指有受控方或共同受控方至少50%直接或间接的投票权、资金或其他有价证券。 1. 授予版权许可 每个“贡献者”根据“本许可证”授予您永久性的、全球性的、免费的、非独占的、不可撤销的版权许可,您可以复制、使用、修改、分发其“贡献”,不论修改与否。 2. 授予专利许可 每个“贡献者”根据“本许可证”授予您永久性的、全球性的、免费的、非独占的、不可撤销的(根据本条规定撤销除外)专利许可,供您制造、委托制造、使用、许诺销售、销售、进口其“贡献”或以其他方式转移其“贡献”。前述专利许可仅限于“贡献者”现在或将来拥有或控制的其“贡献”本身或其“贡献”与许可“贡献”时的“软件”结合而将必然会侵犯的专利权利要求,不包括对“贡献”的修改或包含“贡献”的其他结合。如果您或您的“关联实体”直接或间接地,就“软件”或其中的“贡献”对任何人发起专利侵权诉讼(包括反诉或交叉诉讼)或其他专利维权行动,指控其侵犯专利权,则“本许可证”授予您对“软件”的专利许可自您提起诉讼或发起维权行动之日终止。 3. 无商标许可 “本许可证”不提供对“贡献者”的商品名称、商标、服务标志或产品名称的商标许可,但您为满足第4条规定的声明义务而必须使用除外。 4. 分发限制 您可以在任何媒介中将“软件”以源程序形式或可执行形式重新分发,不论修改与否,但您必须向接收者提供“本许可证”的副本,并保留“软件”中的版权、商标、专利及免责声明。 5. 免责声明与责任限制 “软件”及其中的“贡献”在提供时不带任何明示或默示的担保。在任何情况下,“贡献者”或版权所有者不对任何人因使用“软件”或其中的“贡献”而引发的任何直接或间接损失承担责任,不论因何种原因导致或者基于何种法律理论,即使其曾被建议有此种损失的可能性。 6. 语言 “本许可证”以中英文双语表述,中英文版本具有同等法律效力。如果中英文版本存在任何冲突不一致,以中文版为准。 条款结束 如何将木兰宽松许可证,第2版,应用到您的软件 如果您希望将木兰宽松许可证,第2版,应用到您的新软件,为了方便接收者查阅,建议您完成如下三步: 1, 请您补充如下声明中的空白,包括软件名、软件的首次发表年份以及您作为版权人的名字; 2, 请您在软件包的一级目录下创建以“LICENSE”为名的文件,将整个许可证文本放入该文件中; 3, 请将如下声明文本放入每个源文件的头部注释中。 Copyright (c) [Year] [name of copyright holder] [Software Name] is licensed under Mulan PSL v2. You can use this software according to the terms and conditions of the Mulan PSL v2. You may obtain a copy of Mulan PSL v2 at: http://license.coscl.org.cn/MulanPSL2 THIS SOFTWARE IS PROVIDED ON AN "AS IS" BASIS, WITHOUT WARRANTIES OF ANY KIND, EITHER EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO NON-INFRINGEMENT, MERCHANTABILITY OR FIT FOR A PARTICULAR PURPOSE. See the Mulan PSL v2 for more details. Mulan Permissive Software License,Version 2 Mulan Permissive Software License,Version 2 (Mulan PSL v2) January 2020 http://license.coscl.org.cn/MulanPSL2 Your reproduction, use, modification and distribution of the Software shall be subject to Mulan PSL v2 (this License) with the following terms and conditions: 0. Definition Software means the program and related documents which are licensed under this License and comprise all Contribution(s). Contribution means the copyrightable work licensed by a particular Contributor under this License. Contributor means the Individual or Legal Entity who licenses its copyrightable work under this License. Legal Entity means the entity making a Contribution and all its Affiliates. Affiliates means entities that control, are controlled by, or are under common control with the acting entity under this License, ‘control’ means direct or indirect ownership of at least fifty percent (50%) of the voting power, capital or other securities of controlled or commonly controlled entity. 1. Grant of Copyright License Subject to the terms and conditions of this License, each Contributor hereby grants to you a perpetual, worldwide, royalty-free, non-exclusive, irrevocable copyright license to reproduce, use, modify, or distribute its Contribution, with modification or not. 2. Grant of Patent License Subject to the terms and conditions of this License, each Contributor hereby grants to you a perpetual, worldwide, royalty-free, non-exclusive, irrevocable (except for revocation under this Section) patent license to make, have made, use, offer for sale, sell, import or otherwise transfer its Contribution, where such patent license is only limited to the patent claims owned or controlled by such Contributor now or in future which will be necessarily infringed by its Contribution alone, or by combination of the Contribution with the Software to which the Contribution was contributed. The patent license shall not apply to any modification of the Contribution, and any other combination which includes the Contribution. If you or your Affiliates directly or indirectly institute patent litigation (including a cross claim or counterclaim in a litigation) or other patent enforcement activities against any individual or entity by alleging that the Software or any Contribution in it infringes patents, then any patent license granted to you under this License for the Software shall terminate as of the date such litigation or activity is filed or taken. 3. No Trademark License No trademark license is granted to use the trade names, trademarks, service marks, or product names of Contributor, except as required to fulfill notice requirements in Section 4. 4. Distribution Restriction You may distribute the Software in any medium with or without modification, whether in source or executable forms, provided that you provide recipients with a copy of this License and retain copyright, patent, trademark and disclaimer statements in the Software. 5. Disclaimer of Warranty and Limitation of Liability THE SOFTWARE AND CONTRIBUTION IN IT ARE PROVIDED WITHOUT WARRANTIES OF ANY KIND, EITHER EXPRESS OR IMPLIED. IN NO EVENT SHALL ANY CONTRIBUTOR OR COPYRIGHT HOLDER BE LIABLE TO YOU FOR ANY DAMAGES, INCLUDING, BUT NOT LIMITED TO ANY DIRECT, OR INDIRECT, SPECIAL OR CONSEQUENTIAL DAMAGES ARISING FROM YOUR USE OR INABILITY TO USE THE SOFTWARE OR THE CONTRIBUTION IN IT, NO MATTER HOW IT’S CAUSED OR BASED ON WHICH LEGAL THEORY, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. 6. Language THIS LICENSE IS WRITTEN IN BOTH CHINESE AND ENGLISH, AND THE CHINESE VERSION AND ENGLISH VERSION SHALL HAVE THE SAME LEGAL EFFECT. IN THE CASE OF DIVERGENCE BETWEEN THE CHINESE AND ENGLISH VERSIONS, THE CHINESE VERSION SHALL PREVAIL. END OF THE TERMS AND CONDITIONS How to Apply the Mulan Permissive Software License,Version 2 (Mulan PSL v2) to Your Software To apply the Mulan PSL v2 to your work, for easy identification by recipients, you are suggested to complete following three steps: i Fill in the blanks in following statement, including insert your software name, the year of the first publication of your software, and your name identified as the copyright owner; ii Create a file named “LICENSE” which contains the whole context of this License in the first directory of your software package; iii Attach the statement to the appropriate annotated syntax at the beginning of each source file. Copyright (c) [Year] [name of copyright holder] [Software Name] is licensed under Mulan PSL v2. You can use this software according to the terms and conditions of the Mulan PSL v2. You may obtain a copy of Mulan PSL v2 at: http://license.coscl.org.cn/MulanPSL2 THIS SOFTWARE IS PROVIDED ON AN "AS IS" BASIS, WITHOUT WARRANTIES OF ANY KIND, EITHER EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO NON-INFRINGEMENT, MERCHANTABILITY OR FIT FOR A PARTICULAR PURPOSE. See the Mulan PSL v2 for more details.

简介

基于深度学习的2D虚拟试衣并进行模型压缩优化 展开 收起
Python 等 2 种语言
MulanPSL-2.0
取消

发行版

暂无发行版

贡献者

全部

近期动态

不能加载更多了
马建仓 AI 助手
尝试更多
代码解读
代码找茬
代码优化