matting
June 7, 2025 · View on GitHub
基于PPMatting的抠图
下载模型
-
matting/matting.static.512.onnx: PPMatting.512, Model Type: onnx, input size: 1x3x512x512 download
-
matting/matting.static.1024.onnx: PPMatting.1024, Model Type: onnx, input size: 1x3x1024x1024 download
获取测试图片

运行demo
注:请将PATH更换为自己对应的目录
- --name: 模型名称
- --inference_type: 推理后端类型
- --device_type: 推理后端的执行设备类型
- --model_type: 模型类型
- --is_path: 模型是否为路径
- --model_value: 模型路径或模型文件
- --codec_flag: 编解码类型
- --parallel_type: 并行类型
- --input_path: 输入图片路径
- --output_path: 输出图片路径
- --model_inputs: 模型输入
- --model_outputs: 模型输出
推理后端为onnxruntime,推理执行设备为CUDA
# 进入目录
cd /yourpath/nndeploy/build
# 链接
export LD_LIBRARY_PATH=$(pwd):$LD_LIBRARY_PATH
export LD_LIBRARY_PATH=/home/resource/third_party/onnxruntime-linux-aarch64-1.20.1/lib:$LD_LIBRARY_PATH
# 串行执行
./nndeploy_demo_matting --name nndeploy::matting::ppmatting --inference_type kInferenceTypeOnnxRuntime --device_type kDeviceTypeCodeCuda:0 --model_type kModelTypeOnnx --is_path --codec_flag kCodecFlagImage --parallel_type kParallelTypeSequential --input_path matting_input.jpg --output_path matting_output.jpg --model_value /home/for_all_users/model/segment/pp_matting/matting.static.512.onnx --model_inputs img --model_outputs fetch_name_0
# 耗时
decode_node run() 100 126.854 1.269 0.000
nndeploy::matting::ppmatting run() 100 10875.442 108.754 0.000
preprocess run() 100 68.100 0.681 0.000
infer run() 100 10780.815 107.808 0.000
postprocess run() 100 25.965 0.260 0.000
vis_matting_node run() 100 85.871 0.859 0.000
encode_node run() 100 88.013 0.880 0.000
----------------------------------------------------------------------------------------------
#流水线执行
./nndeploy_demo_matting --name nndeploy::matting::ppmatting --inference_type kInferenceTypeOnnxRuntime --device_type kDeviceTypeCodeCuda:0 --model_type kModelTypeOnnx --is_path --codec_flag kCodecFlagImage --parallel_type kParallelTypePipeline --input_path matting_input.jpg --output_path matting_output.jpg --model_value /home/for_all_users/model/segment/pp_matting/matting.static.512.onnx --model_inputs img --model_outputs fetch_name_0
# 耗时
TimeProfiler: demo
----------------------------------------------------------------------------------------------
name call_times sum cost_time(ms) avg cost_time(ms) gflops
----------------------------------------------------------------------------------------------
graph->init() 1 177.771 177.771 0.000
graph->run 1 11001.096 11001.096 0.000
demo run() 100 0.008 0.000 0.000
decode_node run() 100 10252.853 102.529 0.000
nndeploy::matting::ppmatting run() 100 0.213 0.002 0.000
preprocess run() 100 93.440 0.934 0.000
infer run() 100 10998.621 109.986 0.000
postprocess run() 100 39.947 0.399 0.000
vis_matting_node run() 100 94.022 0.940 0.000
encode_node run() 100 101.140 1.011 0.000
----------------------------------------------------------------------------------------------
效果示例
输入图片

输出图片
